
import tensorflow as tf
tf.config.threading.set_intra_op_parallelism_threads(6)
tf.config.threading.set_inter_op_parallelism_threads(6)

2025-07-27 22:15:33.475126: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
/Users/andrewplattel/opt/anaconda3/lib/python3.9/site-packages/scipy/__init__.py:155: UserWarning: A NumPy version >=1.18.5 and <1.25.0 is required for this version of SciPy (detected version 1.26.4
  warnings.warn(f"A NumPy version >={np_minversion} and <{np_maxversion}"


print("Intra-op:", tf.config.threading.get_intra_op_parallelism_threads())
print("Inter-op:", tf.config.threading.get_inter_op_parallelism_threads())

Intra-op: 6
Inter-op: 6


%matplotlib inline
from tensorflow import keras
from tensorflow.keras.optimizers import Adam
from tensorflow.keras.losses import BinaryCrossentropy
from tensorflow.keras.metrics import BinaryAccuracy
from sklearn.model_selection import train_test_split
from tensorflow.keras import datasets, layers, models
from tensorflow.keras.models import Sequential, Model
from tensorflow.keras.layers import LSTM, Dense, Input, Lambda
from sklearn.preprocessing import MinMaxScaler
from sklearn.metrics import mean_squared_error, mean_absolute_error
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
import scipy.stats
import pandas as pd
import time


rawdata = pd.read_csv('hall.csv')
data = rawdata
data.head()


# Convert 'time' to datetime and set it as index
data['time'] = pd.to_datetime(data['time'])
data.set_index('time', inplace=True)
data.head()


data = data.sort_index() #ensure we order by time
data.head()


#extract the blood glucose variable to train a model on
data_var = data['gl']
data_var.info

# # Normalize the data
scaler = MinMaxScaler(feature_range=(0, 1))
data_var_clean = data_var.dropna()  # Remove NaN values
data_normalized = scaler.fit_transform(data_var_clean.values.reshape(-1, 1))
data_normalized = pd.Series(data_normalized.flatten())
data_normalized

0         0.201521
1         0.201521
2         0.201521
3         0.209125
4         0.212928
            ...   
105412    0.114068
105413    0.091255
105414    0.079848
105415    0.083650
105416    0.098859
Length: 105417, dtype: float64


# Create sequences
def create_sequences(data, seq_length):
    X, y = [], []
    for i in range(len(data) - seq_length):
        X.append(data[i:(i + seq_length)].values)
        y.append(data[i + seq_length])
    return np.array(X), np.array(y)

# Set sequence length (e.g., use x time entries to predict the next)
seq_length = 70

X, y = create_sequences(data_normalized, seq_length)

# Reshape X to be [samples, time steps, features]
X = np.reshape(X, (X.shape[0], X.shape[1], 1))

# Split the data into training and testing sets
train_size = int(len(X) * 0.8)
X_train, X_test = X[:train_size], X[train_size:]
y_train, y_test = y[:train_size], y[train_size:]


# Build the RNN model
def create_model(seq_length_in, seq_length_out):
    inputs = Input(shape=(seq_length_in, 1))
    x = LSTM(50, activation='relu', return_sequences=True)(inputs)
    x = LSTM(50, activation='relu')(x)
    outputs = Dense(seq_length_out)(x)
    model = Model(inputs=inputs, outputs=outputs)
    model.compile(optimizer='adam', loss='mse')
    return model


# Create and compile the model
seq_length_out = 1
model = create_model(seq_length, seq_length_out)


start_time = time.time()
# Train the model (takes time)
history = model.fit(X_train, y_train, epochs=5, batch_size=320, validation_split=0.2, verbose=0)

end_time = time.time()
elapsed_time = end_time - start_time

print(f"Execution time: {elapsed_time} seconds")

Execution time: 186.02215194702148 seconds


start_time = time.time()
# Make predictions
train_predict = model.predict(X_train)
test_predict = model.predict(X_test)
end_time = time.time()
elapsed_time1 = end_time - start_time
print(f"Execution time: {elapsed_time1} seconds")

start_time = time.time()
# Invert predictions to original scale
train_predict = scaler.inverse_transform(train_predict)
y_train_inv = scaler.inverse_transform(y_train.reshape(-1, 1))
test_predict = scaler.inverse_transform(test_predict)
y_test_inv = scaler.inverse_transform(y_test.reshape(-1, 1))
end_time = time.time()
elapsed_time2 = end_time - start_time
print(f"Execution time: {elapsed_time2} seconds")

2634/2634 ━━━━━━━━━━━━━━━━━━━━ 23s 8ms/step
659/659 ━━━━━━━━━━━━━━━━━━━━ 6s 8ms/step
Execution time: 29.487671852111816 seconds
Execution time: 0.00886392593383789 seconds


start_time = time.time()
# Calculate RMSE
train_rmse = np.sqrt(np.mean((train_predict - y_train_inv)**2))
test_rmse = np.sqrt(np.mean((test_predict - y_test_inv)**2))
end_time = time.time()
elapsed_time3 = end_time - start_time
print(f"Execution time: {elapsed_time3} seconds")

print(f"Train RMSE: {train_rmse} mg/DL")
print(f"Test RMSE: {test_rmse} mg/DL")

Execution time: 0.00877690315246582 seconds
Train RMSE: 14.491917918641693 mg/DL
Test RMSE: 13.411277322809406 mg/DL


plt.scatter(y_test_inv, test_predict, label='Predictions')
plt.plot([min(y_test_inv), max(y_test_inv)], [min(y_test_inv), max(y_test_inv)], color='red', linestyle='--', label='y = x')
plt.xlabel('Actual BG')
plt.ylabel('Predicted BG')
plt.legend()
plt.title('Predicted BG vs Actual BG')
plt.show()


ids = data.id.unique().tolist()
ids[0:5]

['1636-69-001', '1636-69-064', '1636-69-028', '1636-69-053', '1636-69-069']


grouped_data = {f'id_{i}_data': group for i, (id_, group) in enumerate(data.groupby('id'))}


print(ids[26])
grouped_data['id_26_data'] #the associated data to the person that id index

2133-004


#Creating normalized data from the dictionary

normalized_data = {}

for key, group in grouped_data.items():
    data_var = group['gl']
    data_var_clean = data_var.dropna()

    if len(data_var_clean) > 0:
        scaler = MinMaxScaler(feature_range=(0, 1))
        data_normalized = scaler.fit_transform(data_var_clean.values.reshape(-1, 1))
        normalized_data[key] = pd.Series(data_normalized.flatten(), index=data_var_clean.index)
    else:
        normalized_data[key] = pd.Series([], dtype='float64')


normalized_data['id_0_data']

time
2014-02-03 03:42:12    0.155914
2014-02-03 03:47:12    0.155914
2014-02-03 03:52:12    0.155914
2014-02-03 03:57:12    0.166667
2014-02-03 04:02:12    0.172043
                         ...   
2015-04-02 14:48:06    0.274194
2015-04-02 14:53:06    0.268817
2015-04-02 14:58:06    0.284946
2015-04-02 15:03:06    0.306452
2015-04-02 15:08:06    0.295699
Length: 1846, dtype: float64


seq_length_in = 70 # number of datapoints to use to predict the output, in this case 70 (or 350 minutes)
seq_length_out = 1 # number of datapoints to predict per iteration, this case just the next 5 minute interval
data_subset = normalized_data['id_0_data']

model_0 = create_model(seq_length, seq_length_out)


X, y = create_sequences(data_subset, seq_length_in)

# Reshape X to be [samples, time steps, features]
X = np.reshape(X, (X.shape[0], X.shape[1], 1))

# Split the data into training and testing sets
train_size = int(len(X) * 0.8)
X_train, X_test = X[:train_size], X[train_size:]
y_train, y_test = y[:train_size], y[train_size:]


start_time = time.time()
# Train the model
history = model_0.fit(X_train, y_train, epochs=5, batch_size = (int(len(data_subset)/100)), validation_split=0.2, verbose=0) #batch size is the number of datapoints for the individual divided by 100, 
                                                                                                                            # rounded to the nearest integer

end_time = time.time()
elapsed_time = end_time - start_time

print(f"Execution time: {elapsed_time} seconds")

Execution time: 10.707152128219604 seconds


start_time = time.time()
# Make predictions
train_predict = model_0.predict(X_train)
test_predict = model_0.predict(X_test)
end_time = time.time()
elapsed_time1 = end_time - start_time
print(f"Execution time: {elapsed_time1} seconds")

start_time = time.time()
# Invert predictions to original scale
train_predict = scaler.inverse_transform(train_predict)
y_train_inv = scaler.inverse_transform(y_train.reshape(-1, 1))
test_predict = scaler.inverse_transform(test_predict)
y_test_inv = scaler.inverse_transform(y_test.reshape(-1, 1))
end_time = time.time()
elapsed_time2 = end_time - start_time
print(f"Execution time: {elapsed_time2} seconds")

45/45 ━━━━━━━━━━━━━━━━━━━━ 1s 13ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 10ms/step
Execution time: 0.9860191345214844 seconds
Execution time: 0.0007960796356201172 seconds


start_time = time.time()
# Calculate RMSE
train_rmse = np.sqrt(np.mean((train_predict - y_train_inv)**2))
test_rmse = np.sqrt(np.mean((test_predict - y_test_inv)**2))
end_time = time.time()
elapsed_time3 = end_time - start_time
print(f"Execution time: {elapsed_time3} seconds")

print(f"Train RMSE: {train_rmse} mg/DL")
print(f"Test RMSE: {test_rmse} mg/DL")

Execution time: 0.0003910064697265625 seconds
Train RMSE: 8.434181792616451 mg/DL
Test RMSE: 6.895865944761983 mg/DL


plt.scatter(y_test_inv, test_predict, label='Predictions')
plt.plot([min(y_test_inv), max(y_test_inv)], [min(y_test_inv), max(y_test_inv)], color='red', linestyle='--', label='y = x')
plt.xlabel('Actual BG')
plt.ylabel('Predicted BG')
plt.legend()
plt.title('Predicted BG vs Actual BG')
plt.show()


def create_sequences(data, seq_length_in, seq_length_out):
    X, y = [], []
    for i in range(len(data) - seq_length_in - seq_length_out + 1):
        X.append(data[i:(i + seq_length_in)].values)
        y.append(data[(i + seq_length_in):(i + seq_length_in + seq_length_out)].values)
    return np.array(X), np.array(y)


# Attempt at making a function that does this all given the subset, seq_in and seq_out

def RNN_indv(data_subset, seq_length_in, seq_length_out):
    """ 
    A model that takes in:\
        data_subset, the data partaining to one id (person from study), 
        seq_length_in,  the number of datapoints to use to predict the output
        seq_length_out, the number of datapoints to predict per iteration (datapoints are 5 min intervals)
        
    Then splits the data into training/testing sets, trains the model, makes predictions based on the seq_length_out,
    calcualtes the RMSE (in the original units), and creates a plot to visualize the accuracy.
    
    """

    model = create_model(seq_length_in, seq_length_out)
    
    #############
    
    X, y = create_sequences(data_subset, seq_length_in, seq_length_out)

    # Reshape X to be [samples, time steps, features]
    X = np.reshape(X, (X.shape[0], X.shape[1], 1))

    # Flatten y to shape (samples, seq_length_out)
    y = y.reshape((y.shape[0], y.shape[1]))


    # Split the data into training and testing sets
    train_size = int(len(X) * 0.8)
    X_train, X_test = X[:train_size], X[train_size:]
    y_train, y_test = y[:train_size], y[train_size:]

    #############
    
    start_time = time.time()
    # Train the model
    history = model.fit(X_train, y_train, epochs=35, batch_size = (int(len(data_subset)/100)), 
                          validation_split=0.2, verbose=0) 
    # batch size is the number of datapoints for the individual divided by 10, rounded to the nearest integer

    end_time = time.time()
    elapsed_time = end_time - start_time

    print(f"Execution time: {elapsed_time} seconds")

    #############
    
    start_time = time.time()
    # Make predictions
    train_predict = model.predict(X_train)
    test_predict = model.predict(X_test)
    end_time = time.time()
    elapsed_time1 = end_time - start_time
    print(f"Execution time: {elapsed_time1} seconds")

    start_time = time.time()
    # Invert predictions to original scale
    train_predict = scaler.inverse_transform(train_predict)
    y_train_inv = scaler.inverse_transform(y_train)
    test_predict = scaler.inverse_transform(test_predict)
    y_test_inv = scaler.inverse_transform(y_test)
    end_time = time.time()
    elapsed_time2 = end_time - start_time
    print(f"Execution time: {elapsed_time2} seconds")

    #############

    start_time = time.time()
    # Calculate RMSE
    train_rmse = np.sqrt(np.mean((train_predict - y_train_inv)**2))
    test_rmse = np.sqrt(np.mean((test_predict - y_test_inv)**2))
    end_time = time.time()
    elapsed_time3 = end_time - start_time
    print(f"Execution time: {elapsed_time3} seconds")
    print("\n")
    print(f"Overall Train RMSE: {train_rmse} mg/DL")
    print(f"Overall Test RMSE: {test_rmse} mg/DL")
    print("\n")
    # Calculate RMSE for each individual forecast timestep
    per_timestep_test_rmse = np.sqrt(np.mean((test_predict - y_test_inv)**2, axis=0))
    per_timestep_train_rmse = np.sqrt(np.mean((train_predict - y_train_inv)**2, axis=0))
   
    # Print the per-timestep train RMSEs
    for i, rmse in enumerate(per_timestep_train_rmse):
        print(f"Train RMSE at t+{i+1}: {rmse:.2f} mg/DL")
    
    print("\n")    
    # Print the per-timestep test RMSEs
    for i, rmse in enumerate(per_timestep_test_rmse):
        print(f"Test RMSE at t+{i+1}: {rmse:.2f} mg/DL")

    #############
    
    # Flatten the predictions and actual values
    y_flat = y_test_inv.flatten()
    pred_flat = test_predict.flatten()

    # Create color labels based on output index
    n_outputs = seq_length_out
    n_sequences = y_test_inv.shape[0]

    # This will repeat [0, 1, ..., n_outputs-1] for each sequence
    color_labels = np.tile(np.arange(n_outputs), n_sequences)

    # Create a colormap
    cmap = plt.get_cmap('viridis', n_outputs)

    # Plot with color mapping
    scatter = plt.scatter(y_flat, pred_flat, c=color_labels, cmap=cmap)
    plt.plot([np.min(y_flat), np.max(y_flat)], [np.min(y_flat), np.max(y_flat)],
             color='red', linestyle='--', label='y = x')

    # Add colorbar with labels
    cbar = plt.colorbar(scatter, ticks=np.arange(n_outputs))
    cbar.set_label('Output Timestep')
    cbar.set_ticklabels([f't+{i+1}' for i in range(n_outputs)])

    plt.xlabel('Actual BG')
    plt.ylabel('Predicted BG')
    plt.title('Predicted BG vs Actual BG (Colored by Forecast Horizon)')
    plt.legend()
    plt.show()
    
    return test_predict, y_test_inv


seq_length_in = 12 
seq_length_out = 6 
data_subset = normalized_data['id_20_data']

RNN_indv(data_subset, seq_length_in, seq_length_out)

Execution time: 14.721097946166992 seconds
46/46 ━━━━━━━━━━━━━━━━━━━━ 1s 7ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.6433248519897461 seconds
Execution time: 0.0005776882171630859 seconds
Execution time: 0.00016307830810546875 seconds


Overall Train RMSE: 9.697396266139112 mg/DL
Overall Test RMSE: 14.244852198045646 mg/DL


Train RMSE at t+1: 4.82 mg/DL
Train RMSE at t+2: 6.74 mg/DL
Train RMSE at t+3: 8.58 mg/DL
Train RMSE at t+4: 10.57 mg/DL
Train RMSE at t+5: 11.85 mg/DL
Train RMSE at t+6: 13.02 mg/DL


Test RMSE at t+1: 5.25 mg/DL
Test RMSE at t+2: 8.18 mg/DL
Test RMSE at t+3: 11.47 mg/DL
Test RMSE at t+4: 14.81 mg/DL
Test RMSE at t+5: 17.75 mg/DL
Test RMSE at t+6: 21.38 mg/DL

(array([[114.25487 , 113.75629 , 112.63672 , 112.30811 , 109.97957 ,
         110.85521 ],
        [114.224396, 113.70724 , 112.53086 , 112.204185, 109.959694,
         110.83667 ],
        [115.242836, 114.76751 , 113.51902 , 113.10954 , 110.94312 ,
         111.684425],
        ...,
        [ 90.79295 ,  91.12169 ,  91.876015,  93.95044 ,  93.41583 ,
          95.55196 ],
        [ 88.90803 ,  89.130455,  89.84357 ,  92.19615 ,  91.65675 ,
          94.02229 ],
        [ 89.110825,  89.376175,  89.952644,  92.33615 ,  91.8114  ,
          94.083496]], dtype=float32),
 array([[112.02312139, 112.90751445, 113.79190751, 113.79190751,
         113.79190751, 114.67630058],
        [112.90751445, 113.79190751, 113.79190751, 113.79190751,
         114.67630058, 115.56069364],
        [113.79190751, 113.79190751, 113.79190751, 114.67630058,
         115.56069364, 115.56069364],
        ...,
        [ 84.60693642,  84.60693642,  84.60693642,  83.72254335,
          84.60693642,  83.72254335],
        [ 84.60693642,  84.60693642,  83.72254335,  84.60693642,
          83.72254335,  82.83815029],
        [ 84.60693642,  83.72254335,  84.60693642,  83.72254335,
          82.83815029,  85.49132948]]))


all_preds1 = []
all_truths1 = []
seq_length_in = 70 # Training on 350 mins of data (5.83 hours)
seq_length_out = 6 # 30 mins (~5 min interval each interation)
results1 = {}

for key, subset in normalized_data.items():
    print(f"\nRunning model for: {key}")
    
    try:
        preds, truths = RNN_indv(subset, seq_length_in, seq_length_out)
        all_preds1.append(preds)
        all_truths1.append(truths)
        results1[key] = (preds, truths)
    except Exception as e:
        print(f"Error for {key}: {e}")
        results1[key] = None

Running model for: id_0_data
Execution time: 58.564422845840454 seconds
45/45 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 9ms/step
Execution time: 0.8954308032989502 seconds
Execution time: 0.0005500316619873047 seconds
Execution time: 0.0001220703125 seconds


Overall Train RMSE: 9.900586240885847 mg/DL
Overall Test RMSE: 7.094973716375761 mg/DL


Train RMSE at t+1: 4.33 mg/DL
Train RMSE at t+2: 6.38 mg/DL
Train RMSE at t+3: 8.48 mg/DL
Train RMSE at t+4: 9.91 mg/DL
Train RMSE at t+5: 12.46 mg/DL
Train RMSE at t+6: 14.26 mg/DL


Test RMSE at t+1: 3.12 mg/DL
Test RMSE at t+2: 4.47 mg/DL
Test RMSE at t+3: 6.16 mg/DL
Test RMSE at t+4: 7.16 mg/DL
Test RMSE at t+5: 8.85 mg/DL
Test RMSE at t+6: 10.23 mg/DL

Running model for: id_1_data
Execution time: 56.83130979537964 seconds
43/43 ━━━━━━━━━━━━━━━━━━━━ 1s 8ms/step
11/11 ━━━━━━━━━━━━━━━━━━━━ 1s 51ms/step
Execution time: 1.1245758533477783 seconds
Execution time: 0.0006070137023925781 seconds
Execution time: 0.00011086463928222656 seconds


Overall Train RMSE: 12.763975172483198 mg/DL
Overall Test RMSE: 18.826685515109748 mg/DL


Train RMSE at t+1: 5.13 mg/DL
Train RMSE at t+2: 7.88 mg/DL
Train RMSE at t+3: 10.68 mg/DL
Train RMSE at t+4: 13.66 mg/DL
Train RMSE at t+5: 16.14 mg/DL
Train RMSE at t+6: 18.11 mg/DL


Test RMSE at t+1: 11.18 mg/DL
Test RMSE at t+2: 14.98 mg/DL
Test RMSE at t+3: 18.17 mg/DL
Test RMSE at t+4: 20.69 mg/DL
Test RMSE at t+5: 22.04 mg/DL
Test RMSE at t+6: 23.10 mg/DL

Running model for: id_2_data
Execution time: 55.44075894355774 seconds
44/44 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step
11/11 ━━━━━━━━━━━━━━━━━━━━ 0s 10ms/step
Execution time: 0.8859801292419434 seconds
Execution time: 0.0005838871002197266 seconds
Execution time: 8.392333984375e-05 seconds


Overall Train RMSE: 10.050142051108038 mg/DL
Overall Test RMSE: 19.782857751978163 mg/DL


Train RMSE at t+1: 7.71 mg/DL
Train RMSE at t+2: 8.90 mg/DL
Train RMSE at t+3: 10.08 mg/DL
Train RMSE at t+4: 10.75 mg/DL
Train RMSE at t+5: 11.05 mg/DL
Train RMSE at t+6: 11.32 mg/DL


Test RMSE at t+1: 14.39 mg/DL
Test RMSE at t+2: 17.10 mg/DL
Test RMSE at t+3: 19.69 mg/DL
Test RMSE at t+4: 21.25 mg/DL
Test RMSE at t+5: 22.10 mg/DL
Test RMSE at t+6: 22.83 mg/DL

Running model for: id_3_data
Execution time: 65.73585629463196 seconds
43/43 ━━━━━━━━━━━━━━━━━━━━ 1s 14ms/step
11/11 ━━━━━━━━━━━━━━━━━━━━ 0s 10ms/step
Execution time: 1.020150899887085 seconds
Execution time: 0.0006010532379150391 seconds
Execution time: 8.20159912109375e-05 seconds


Overall Train RMSE: 9.830644029698883 mg/DL
Overall Test RMSE: 9.547988060447478 mg/DL


Train RMSE at t+1: 5.77 mg/DL
Train RMSE at t+2: 7.62 mg/DL
Train RMSE at t+3: 9.35 mg/DL
Train RMSE at t+4: 10.66 mg/DL
Train RMSE at t+5: 11.69 mg/DL
Train RMSE at t+6: 12.28 mg/DL


Test RMSE at t+1: 5.74 mg/DL
Test RMSE at t+2: 7.60 mg/DL
Test RMSE at t+3: 9.05 mg/DL
Test RMSE at t+4: 10.16 mg/DL
Test RMSE at t+5: 11.31 mg/DL
Test RMSE at t+6: 11.97 mg/DL

Running model for: id_4_data
Execution time: 61.62102389335632 seconds
53/53 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step
14/14 ━━━━━━━━━━━━━━━━━━━━ 0s 9ms/step
Execution time: 1.0203120708465576 seconds
Execution time: 0.0011529922485351562 seconds
Execution time: 0.00026607513427734375 seconds


Overall Train RMSE: 11.099187725845745 mg/DL
Overall Test RMSE: 10.01048290850761 mg/DL


Train RMSE at t+1: 5.27 mg/DL
Train RMSE at t+2: 7.91 mg/DL
Train RMSE at t+3: 9.85 mg/DL
Train RMSE at t+4: 11.91 mg/DL
Train RMSE at t+5: 13.61 mg/DL
Train RMSE at t+6: 14.99 mg/DL


Test RMSE at t+1: 4.38 mg/DL
Test RMSE at t+2: 6.73 mg/DL
Test RMSE at t+3: 8.81 mg/DL
Test RMSE at t+4: 10.91 mg/DL
Test RMSE at t+5: 12.22 mg/DL
Test RMSE at t+6: 13.81 mg/DL

Running model for: id_5_data
Execution time: 61.30432105064392 seconds
43/43 ━━━━━━━━━━━━━━━━━━━━ 1s 14ms/step
11/11 ━━━━━━━━━━━━━━━━━━━━ 0s 9ms/step
Execution time: 0.9626240730285645 seconds
Execution time: 0.0005691051483154297 seconds
Execution time: 8.916854858398438e-05 seconds


Overall Train RMSE: 15.212611968616024 mg/DL
Overall Test RMSE: 12.389760582840259 mg/DL


Train RMSE at t+1: 7.02 mg/DL
Train RMSE at t+2: 10.51 mg/DL
Train RMSE at t+3: 13.91 mg/DL
Train RMSE at t+4: 16.55 mg/DL
Train RMSE at t+5: 18.79 mg/DL
Train RMSE at t+6: 20.21 mg/DL


Test RMSE at t+1: 6.73 mg/DL
Test RMSE at t+2: 9.14 mg/DL
Test RMSE at t+3: 11.67 mg/DL
Test RMSE at t+4: 13.00 mg/DL
Test RMSE at t+5: 15.20 mg/DL
Test RMSE at t+6: 16.00 mg/DL

Running model for: id_6_data
Execution time: 59.82891082763672 seconds
45/45 ━━━━━━━━━━━━━━━━━━━━ 1s 15ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 10ms/step
Execution time: 1.0510690212249756 seconds
Execution time: 0.0005781650543212891 seconds
Execution time: 8.416175842285156e-05 seconds


Overall Train RMSE: 11.464505910366112 mg/DL
Overall Test RMSE: 12.331822536894423 mg/DL


Train RMSE at t+1: 8.05 mg/DL
Train RMSE at t+2: 9.60 mg/DL
Train RMSE at t+3: 10.55 mg/DL
Train RMSE at t+4: 12.07 mg/DL
Train RMSE at t+5: 12.86 mg/DL
Train RMSE at t+6: 14.47 mg/DL


Test RMSE at t+1: 9.31 mg/DL
Test RMSE at t+2: 10.58 mg/DL
Test RMSE at t+3: 12.16 mg/DL
Test RMSE at t+4: 13.04 mg/DL
Test RMSE at t+5: 13.53 mg/DL
Test RMSE at t+6: 14.59 mg/DL

Running model for: id_7_data
Execution time: 57.23943591117859 seconds
44/44 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step
11/11 ━━━━━━━━━━━━━━━━━━━━ 0s 10ms/step
Execution time: 0.8898119926452637 seconds
Execution time: 0.0005960464477539062 seconds
Execution time: 8.20159912109375e-05 seconds


Overall Train RMSE: 12.710189792336731 mg/DL
Overall Test RMSE: 10.596413650125424 mg/DL


Train RMSE at t+1: 6.42 mg/DL
Train RMSE at t+2: 9.12 mg/DL
Train RMSE at t+3: 11.71 mg/DL
Train RMSE at t+4: 13.80 mg/DL
Train RMSE at t+5: 15.40 mg/DL
Train RMSE at t+6: 16.74 mg/DL


Test RMSE at t+1: 5.64 mg/DL
Test RMSE at t+2: 6.99 mg/DL
Test RMSE at t+3: 9.52 mg/DL
Test RMSE at t+4: 11.64 mg/DL
Test RMSE at t+5: 13.11 mg/DL
Test RMSE at t+6: 13.97 mg/DL

Running model for: id_8_data
Execution time: 56.2462592124939 seconds
38/38 ━━━━━━━━━━━━━━━━━━━━ 1s 13ms/step
10/10 ━━━━━━━━━━━━━━━━━━━━ 0s 10ms/step
Execution time: 0.8301072120666504 seconds
Execution time: 0.0006041526794433594 seconds
Execution time: 7.390975952148438e-05 seconds


Overall Train RMSE: 10.879406695641752 mg/DL
Overall Test RMSE: 9.675346143563461 mg/DL


Train RMSE at t+1: 4.85 mg/DL
Train RMSE at t+2: 7.31 mg/DL
Train RMSE at t+3: 9.72 mg/DL
Train RMSE at t+4: 11.71 mg/DL
Train RMSE at t+5: 13.52 mg/DL
Train RMSE at t+6: 14.79 mg/DL


Test RMSE at t+1: 4.41 mg/DL
Test RMSE at t+2: 6.67 mg/DL
Test RMSE at t+3: 8.78 mg/DL
Test RMSE at t+4: 10.60 mg/DL
Test RMSE at t+5: 11.82 mg/DL
Test RMSE at t+6: 12.98 mg/DL

Running model for: id_9_data
Execution time: 56.92694091796875 seconds
46/46 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 9ms/step
Execution time: 0.8943428993225098 seconds
Execution time: 0.0005619525909423828 seconds
Execution time: 9.012222290039062e-05 seconds


Overall Train RMSE: 11.848520758183172 mg/DL
Overall Test RMSE: 14.557583356932046 mg/DL


Train RMSE at t+1: 5.65 mg/DL
Train RMSE at t+2: 8.37 mg/DL
Train RMSE at t+3: 10.36 mg/DL
Train RMSE at t+4: 12.57 mg/DL
Train RMSE at t+5: 14.53 mg/DL
Train RMSE at t+6: 16.25 mg/DL


Test RMSE at t+1: 7.95 mg/DL
Test RMSE at t+2: 11.12 mg/DL
Test RMSE at t+3: 13.52 mg/DL
Test RMSE at t+4: 15.36 mg/DL
Test RMSE at t+5: 17.27 mg/DL
Test RMSE at t+6: 19.18 mg/DL


# Concatenate all predictions and truths
all_preds_concat1 = np.concatenate(all_preds1, axis=0).flatten()
all_truths_concat1 = np.concatenate(all_truths1, axis=0).flatten()

# Compute overall RMSE and MAE
combined_rmse1 = np.sqrt(mean_squared_error(all_truths_concat1, all_preds_concat1))
combined_mae1 = mean_absolute_error(all_truths_concat1, all_preds_concat1)

print(f"\nCombined RMSE across all IDs: {combined_rmse1:.2f} mg/dL")
print(f"Combined MAE  across all IDs: {combined_mae1:.2f} mg/dL\n")

# Lists to store RMSE and MAE per individual
rmse_per_individual = []
mae_per_individual = []

# Compute per-ID RMSE and MAE
for pred_array, true_array in zip(all_preds1, all_truths1):
    pred_flat = pred_array.flatten()
    true_flat = true_array.flatten()
    
    rmse = np.sqrt(mean_squared_error(true_flat, pred_flat))
    mae = mean_absolute_error(true_flat, pred_flat)
    
    rmse_per_individual.append(rmse)
    mae_per_individual.append(mae)

# Print all 56 RMSEs and MAEs
for i, (rmse, mae) in enumerate(zip(rmse_per_individual, mae_per_individual)):
    print(f"ID {i:2d}: RMSE = {rmse:.2f} mg/dL | MAE = {mae:.2f} mg/dL")

# Summary: min/max RMSE and MAE
print('\n')
print(f'The min RMSE is for ID {np.argmin(rmse_per_individual)} with a value of: {min(rmse_per_individual):.2f}')
print(f'The max RMSE is for ID {np.argmax(rmse_per_individual)} with a value of: {max(rmse_per_individual):.2f}')
print(f'The min MAE  is for ID {np.argmin(mae_per_individual)} with a value of: {min(mae_per_individual):.2f}')
print(f'The max MAE  is for ID {np.argmax(mae_per_individual)} with a value of: {max(mae_per_individual):.2f}')

Combined RMSE across all IDs: 12.82 mg/dL
Combined MAE  across all IDs: 8.45 mg/dL

ID  0: RMSE = 7.09 mg/dL | MAE = 4.68 mg/dL
ID  1: RMSE = 18.83 mg/dL | MAE = 11.36 mg/dL
ID  2: RMSE = 19.78 mg/dL | MAE = 12.41 mg/dL
ID  3: RMSE = 9.55 mg/dL | MAE = 7.32 mg/dL
ID  4: RMSE = 10.01 mg/dL | MAE = 6.64 mg/dL
ID  5: RMSE = 12.39 mg/dL | MAE = 9.24 mg/dL
ID  6: RMSE = 12.33 mg/dL | MAE = 9.11 mg/dL
ID  7: RMSE = 10.60 mg/dL | MAE = 7.16 mg/dL
ID  8: RMSE = 9.68 mg/dL | MAE = 6.91 mg/dL
ID  9: RMSE = 14.56 mg/dL | MAE = 9.36 mg/dL
ID 10: RMSE = 10.43 mg/dL | MAE = 6.81 mg/dL
ID 11: RMSE = 16.00 mg/dL | MAE = 10.36 mg/dL
ID 12: RMSE = 13.54 mg/dL | MAE = 10.48 mg/dL
ID 13: RMSE = 14.48 mg/dL | MAE = 10.58 mg/dL
ID 14: RMSE = 12.37 mg/dL | MAE = 9.01 mg/dL
ID 15: RMSE = 13.86 mg/dL | MAE = 10.10 mg/dL
ID 16: RMSE = 13.08 mg/dL | MAE = 8.50 mg/dL
ID 17: RMSE = 16.41 mg/dL | MAE = 12.10 mg/dL
ID 18: RMSE = 15.98 mg/dL | MAE = 10.74 mg/dL
ID 19: RMSE = 12.26 mg/dL | MAE = 8.69 mg/dL
ID 20: RMSE = 18.14 mg/dL | MAE = 10.57 mg/dL
ID 21: RMSE = 12.16 mg/dL | MAE = 7.84 mg/dL
ID 22: RMSE = 11.23 mg/dL | MAE = 8.76 mg/dL
ID 23: RMSE = 15.75 mg/dL | MAE = 8.99 mg/dL
ID 24: RMSE = 15.20 mg/dL | MAE = 11.10 mg/dL
ID 25: RMSE = 11.36 mg/dL | MAE = 7.43 mg/dL
ID 26: RMSE = 5.88 mg/dL | MAE = 4.23 mg/dL
ID 27: RMSE = 14.41 mg/dL | MAE = 9.09 mg/dL
ID 28: RMSE = 10.43 mg/dL | MAE = 6.72 mg/dL
ID 29: RMSE = 15.85 mg/dL | MAE = 10.41 mg/dL
ID 30: RMSE = 12.02 mg/dL | MAE = 8.17 mg/dL
ID 31: RMSE = 11.92 mg/dL | MAE = 8.31 mg/dL
ID 32: RMSE = 13.66 mg/dL | MAE = 8.10 mg/dL
ID 33: RMSE = 14.47 mg/dL | MAE = 9.97 mg/dL
ID 34: RMSE = 14.49 mg/dL | MAE = 7.33 mg/dL
ID 35: RMSE = 6.95 mg/dL | MAE = 4.57 mg/dL
ID 36: RMSE = 11.27 mg/dL | MAE = 7.70 mg/dL
ID 37: RMSE = 11.26 mg/dL | MAE = 6.49 mg/dL
ID 38: RMSE = 10.81 mg/dL | MAE = 8.58 mg/dL
ID 39: RMSE = 14.75 mg/dL | MAE = 10.52 mg/dL
ID 40: RMSE = 13.07 mg/dL | MAE = 8.38 mg/dL
ID 41: RMSE = 11.23 mg/dL | MAE = 7.41 mg/dL
ID 42: RMSE = 13.64 mg/dL | MAE = 9.40 mg/dL
ID 43: RMSE = 10.01 mg/dL | MAE = 6.76 mg/dL
ID 44: RMSE = 7.75 mg/dL | MAE = 5.50 mg/dL
ID 45: RMSE = 11.21 mg/dL | MAE = 8.65 mg/dL
ID 46: RMSE = 10.81 mg/dL | MAE = 7.21 mg/dL
ID 47: RMSE = 9.49 mg/dL | MAE = 5.82 mg/dL
ID 48: RMSE = 8.96 mg/dL | MAE = 5.84 mg/dL
ID 49: RMSE = 15.70 mg/dL | MAE = 10.10 mg/dL
ID 50: RMSE = 13.00 mg/dL | MAE = 8.76 mg/dL
ID 51: RMSE = 10.95 mg/dL | MAE = 7.65 mg/dL
ID 52: RMSE = 13.49 mg/dL | MAE = 9.70 mg/dL
ID 53: RMSE = 12.35 mg/dL | MAE = 8.78 mg/dL
ID 54: RMSE = 12.51 mg/dL | MAE = 8.87 mg/dL
ID 55: RMSE = 10.03 mg/dL | MAE = 6.15 mg/dL
ID 56: RMSE = 12.69 mg/dL | MAE = 9.42 mg/dL


The min RMSE is for ID 26 with a value of: 5.88
The max RMSE is for ID 2 with a value of: 19.78
The min MAE  is for ID 26 with a value of: 4.23
The max MAE  is for ID 2 with a value of: 12.41


all_preds2 = []
all_truths2 = []
seq_length_in = 60 # Training on 300 mins of data (5 hours)
seq_length_out = 6 # 30 mins (~5 min interval each interation)
results2 = {}

for key, subset in normalized_data.items():
    print(f"\nRunning model for: {key}")
    
    try:
        preds, truths = RNN_indv(subset, seq_length_in, seq_length_out)
        all_preds2.append(preds)
        all_truths2.append(truths)
        results2[key] = (preds, truths)
    except Exception as e:
        print(f"Error for {key}: {e}")
        results2[key] = None

Running model for: id_0_data
Execution time: 48.54973793029785 seconds
45/45 ━━━━━━━━━━━━━━━━━━━━ 1s 11ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 8ms/step
Execution time: 0.837116003036499 seconds
Execution time: 0.0005700588226318359 seconds
Execution time: 7.987022399902344e-05 seconds


Overall Train RMSE: 9.134432471166514 mg/DL
Overall Test RMSE: 7.0869058543926835 mg/DL


Train RMSE at t+1: 4.07 mg/DL
Train RMSE at t+2: 5.38 mg/DL
Train RMSE at t+3: 7.18 mg/DL
Train RMSE at t+4: 9.41 mg/DL
Train RMSE at t+5: 11.86 mg/DL
Train RMSE at t+6: 13.20 mg/DL


Test RMSE at t+1: 3.36 mg/DL
Test RMSE at t+2: 4.45 mg/DL
Test RMSE at t+3: 5.83 mg/DL
Test RMSE at t+4: 7.52 mg/DL
Test RMSE at t+5: 8.83 mg/DL
Test RMSE at t+6: 10.09 mg/DL

Running model for: id_1_data
Execution time: 49.47828507423401 seconds
44/44 ━━━━━━━━━━━━━━━━━━━━ 1s 11ms/step
11/11 ━━━━━━━━━━━━━━━━━━━━ 0s 8ms/step
Execution time: 0.8188338279724121 seconds
Execution time: 0.0005872249603271484 seconds
Execution time: 0.00013899803161621094 seconds


Overall Train RMSE: 12.27185479776481 mg/DL
Overall Test RMSE: 18.187133051950436 mg/DL


Train RMSE at t+1: 5.29 mg/DL
Train RMSE at t+2: 7.68 mg/DL
Train RMSE at t+3: 10.55 mg/DL
Train RMSE at t+4: 12.79 mg/DL
Train RMSE at t+5: 15.55 mg/DL
Train RMSE at t+6: 17.32 mg/DL


Test RMSE at t+1: 10.42 mg/DL
Test RMSE at t+2: 14.83 mg/DL
Test RMSE at t+3: 18.05 mg/DL
Test RMSE at t+4: 19.84 mg/DL
Test RMSE at t+5: 21.52 mg/DL
Test RMSE at t+6: 21.77 mg/DL

Running model for: id_2_data
Execution time: 48.546815156936646 seconds
44/44 ━━━━━━━━━━━━━━━━━━━━ 1s 11ms/step
11/11 ━━━━━━━━━━━━━━━━━━━━ 0s 8ms/step
Execution time: 0.8245770931243896 seconds
Execution time: 0.0005886554718017578 seconds
Execution time: 8.511543273925781e-05 seconds


Overall Train RMSE: 8.122246065713522 mg/DL
Overall Test RMSE: 17.154794426069156 mg/DL


Train RMSE at t+1: 4.17 mg/DL
Train RMSE at t+2: 6.01 mg/DL
Train RMSE at t+3: 7.59 mg/DL
Train RMSE at t+4: 8.89 mg/DL
Train RMSE at t+5: 9.94 mg/DL
Train RMSE at t+6: 10.34 mg/DL


Test RMSE at t+1: 9.72 mg/DL
Test RMSE at t+2: 13.60 mg/DL
Test RMSE at t+3: 16.22 mg/DL
Test RMSE at t+4: 18.65 mg/DL
Test RMSE at t+5: 20.12 mg/DL
Test RMSE at t+6: 21.69 mg/DL

Running model for: id_3_data
Execution time: 50.29539895057678 seconds
43/43 ━━━━━━━━━━━━━━━━━━━━ 1s 11ms/step
11/11 ━━━━━━━━━━━━━━━━━━━━ 0s 8ms/step
Execution time: 0.8213679790496826 seconds
Execution time: 0.0005462169647216797 seconds
Execution time: 0.00012230873107910156 seconds


Overall Train RMSE: 9.964604521093033 mg/DL
Overall Test RMSE: 10.667214834782124 mg/DL


Train RMSE at t+1: 6.23 mg/DL
Train RMSE at t+2: 8.16 mg/DL
Train RMSE at t+3: 9.66 mg/DL
Train RMSE at t+4: 10.67 mg/DL
Train RMSE at t+5: 11.49 mg/DL
Train RMSE at t+6: 12.29 mg/DL


Test RMSE at t+1: 6.17 mg/DL
Test RMSE at t+2: 8.27 mg/DL
Test RMSE at t+3: 10.08 mg/DL
Test RMSE at t+4: 11.45 mg/DL
Test RMSE at t+5: 12.63 mg/DL
Test RMSE at t+6: 13.56 mg/DL

Running model for: id_4_data
Execution time: 50.75205898284912 seconds
53/53 ━━━━━━━━━━━━━━━━━━━━ 1s 10ms/step
14/14 ━━━━━━━━━━━━━━━━━━━━ 0s 8ms/step
Execution time: 0.9146008491516113 seconds
Execution time: 0.0006098747253417969 seconds
Execution time: 8.606910705566406e-05 seconds


Overall Train RMSE: 11.191476426285238 mg/DL
Overall Test RMSE: 9.814197030782438 mg/DL


Train RMSE at t+1: 5.14 mg/DL
Train RMSE at t+2: 7.70 mg/DL
Train RMSE at t+3: 10.11 mg/DL
Train RMSE at t+4: 12.12 mg/DL
Train RMSE at t+5: 13.68 mg/DL
Train RMSE at t+6: 15.15 mg/DL


Test RMSE at t+1: 4.54 mg/DL
Test RMSE at t+2: 7.08 mg/DL
Test RMSE at t+3: 8.69 mg/DL
Test RMSE at t+4: 10.18 mg/DL
Test RMSE at t+5: 12.22 mg/DL
Test RMSE at t+6: 13.36 mg/DL

Running model for: id_5_data
Execution time: 48.85940098762512 seconds
43/43 ━━━━━━━━━━━━━━━━━━━━ 1s 11ms/step
11/11 ━━━━━━━━━━━━━━━━━━━━ 0s 8ms/step
Execution time: 0.8124120235443115 seconds
Execution time: 0.0005612373352050781 seconds
Execution time: 0.00012087821960449219 seconds


Overall Train RMSE: 16.578617343416628 mg/DL
Overall Test RMSE: 12.82396750517587 mg/DL


Train RMSE at t+1: 11.97 mg/DL
Train RMSE at t+2: 13.50 mg/DL
Train RMSE at t+3: 15.28 mg/DL
Train RMSE at t+4: 17.85 mg/DL
Train RMSE at t+5: 18.93 mg/DL
Train RMSE at t+6: 20.33 mg/DL


Test RMSE at t+1: 11.29 mg/DL
Test RMSE at t+2: 11.64 mg/DL
Test RMSE at t+3: 12.36 mg/DL
Test RMSE at t+4: 13.52 mg/DL
Test RMSE at t+5: 13.42 mg/DL
Test RMSE at t+6: 14.42 mg/DL

Running model for: id_6_data
Execution time: 39.110739946365356 seconds
46/46 ━━━━━━━━━━━━━━━━━━━━ 1s 11ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 8ms/step
Execution time: 0.856666088104248 seconds
Execution time: 0.0005910396575927734 seconds
Execution time: 0.00012111663818359375 seconds


Overall Train RMSE: 9.414069673629955 mg/DL
Overall Test RMSE: 11.268502290753512 mg/DL


Train RMSE at t+1: 4.80 mg/DL
Train RMSE at t+2: 6.45 mg/DL
Train RMSE at t+3: 7.94 mg/DL
Train RMSE at t+4: 9.78 mg/DL
Train RMSE at t+5: 11.75 mg/DL
Train RMSE at t+6: 13.06 mg/DL


Test RMSE at t+1: 6.87 mg/DL
Test RMSE at t+2: 8.90 mg/DL
Test RMSE at t+3: 10.33 mg/DL
Test RMSE at t+4: 11.45 mg/DL
Test RMSE at t+5: 13.84 mg/DL
Test RMSE at t+6: 14.36 mg/DL

Running model for: id_7_data
Execution time: 47.84198188781738 seconds
44/44 ━━━━━━━━━━━━━━━━━━━━ 1s 11ms/step
11/11 ━━━━━━━━━━━━━━━━━━━━ 0s 8ms/step
Execution time: 0.8232450485229492 seconds
Execution time: 0.0005371570587158203 seconds
Execution time: 8.273124694824219e-05 seconds


Overall Train RMSE: 12.796524695685331 mg/DL
Overall Test RMSE: 11.925709307415596 mg/DL


Train RMSE at t+1: 6.04 mg/DL
Train RMSE at t+2: 9.25 mg/DL
Train RMSE at t+3: 11.75 mg/DL
Train RMSE at t+4: 14.11 mg/DL
Train RMSE at t+5: 15.58 mg/DL
Train RMSE at t+6: 16.75 mg/DL


Test RMSE at t+1: 4.81 mg/DL
Test RMSE at t+2: 8.27 mg/DL
Test RMSE at t+3: 10.59 mg/DL
Test RMSE at t+4: 13.44 mg/DL
Test RMSE at t+5: 14.83 mg/DL
Test RMSE at t+6: 15.78 mg/DL

Running model for: id_8_data
Execution time: 48.19960808753967 seconds
38/38 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step
10/10 ━━━━━━━━━━━━━━━━━━━━ 0s 8ms/step
Execution time: 0.7692480087280273 seconds
Execution time: 0.0006299018859863281 seconds
Execution time: 7.796287536621094e-05 seconds


Overall Train RMSE: 10.665960559705606 mg/DL
Overall Test RMSE: 9.566320213940225 mg/DL


Train RMSE at t+1: 4.30 mg/DL
Train RMSE at t+2: 7.23 mg/DL
Train RMSE at t+3: 9.40 mg/DL
Train RMSE at t+4: 11.62 mg/DL
Train RMSE at t+5: 13.25 mg/DL
Train RMSE at t+6: 14.60 mg/DL


Test RMSE at t+1: 3.94 mg/DL
Test RMSE at t+2: 6.62 mg/DL
Test RMSE at t+3: 8.62 mg/DL
Test RMSE at t+4: 10.50 mg/DL
Test RMSE at t+5: 11.83 mg/DL
Test RMSE at t+6: 12.85 mg/DL

Running model for: id_9_data
Execution time: 50.49827790260315 seconds
46/46 ━━━━━━━━━━━━━━━━━━━━ 1s 11ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 8ms/step
Execution time: 0.8372890949249268 seconds
Execution time: 0.0006017684936523438 seconds
Execution time: 0.0001652240753173828 seconds


Overall Train RMSE: 11.939792713730121 mg/DL
Overall Test RMSE: 14.88718070977394 mg/DL


Train RMSE at t+1: 5.12 mg/DL
Train RMSE at t+2: 8.12 mg/DL
Train RMSE at t+3: 10.49 mg/DL
Train RMSE at t+4: 12.59 mg/DL
Train RMSE at t+5: 15.04 mg/DL
Train RMSE at t+6: 16.38 mg/DL


Test RMSE at t+1: 7.65 mg/DL
Test RMSE at t+2: 10.95 mg/DL
Test RMSE at t+3: 13.79 mg/DL
Test RMSE at t+4: 15.67 mg/DL
Test RMSE at t+5: 18.17 mg/DL
Test RMSE at t+6: 19.64 mg/DL


# Concatenate all predictions and truths
all_preds_concat2 = np.concatenate(all_preds2, axis=0).flatten()
all_truths_concat2 = np.concatenate(all_truths2, axis=0).flatten()

# Compute overall RMSE and MAE
combined_rmse2 = np.sqrt(mean_squared_error(all_truths_concat2, all_preds_concat2))
combined_mae2 = mean_absolute_error(all_truths_concat2, all_preds_concat2)

print(f"\nCombined RMSE across all IDs: {combined_rmse2:.2f} mg/dL")
print(f"Combined MAE  across all IDs: {combined_mae2:.2f} mg/dL\n")

# Lists to store RMSE and MAE per individual
rmse_per_individual = []
mae_per_individual = []

# Compute per-ID RMSE and MAE
for pred_array, true_array in zip(all_preds2, all_truths2):
    pred_flat = pred_array.flatten()
    true_flat = true_array.flatten()
    
    rmse = np.sqrt(mean_squared_error(true_flat, pred_flat))
    mae = mean_absolute_error(true_flat, pred_flat)
    
    rmse_per_individual.append(rmse)
    mae_per_individual.append(mae)

# Print all 56 RMSEs and MAEs
for i, (rmse, mae) in enumerate(zip(rmse_per_individual, mae_per_individual)):
    print(f"ID {i:2d}: RMSE = {rmse:.2f} mg/dL | MAE = {mae:.2f} mg/dL")

# Summary: min/max RMSE and MAE
print('\n')
print(f'The min RMSE is for ID {np.argmin(rmse_per_individual)} with a value of: {min(rmse_per_individual):.2f}')
print(f'The max RMSE is for ID {np.argmax(rmse_per_individual)} with a value of: {max(rmse_per_individual):.2f}')
print(f'The min MAE  is for ID {np.argmin(mae_per_individual)} with a value of: {min(mae_per_individual):.2f}')
print(f'The max MAE  is for ID {np.argmax(mae_per_individual)} with a value of: {max(mae_per_individual):.2f}')

Combined RMSE across all IDs: 12.70 mg/dL
Combined MAE  across all IDs: 8.32 mg/dL

ID  0: RMSE = 7.09 mg/dL | MAE = 4.86 mg/dL
ID  1: RMSE = 18.19 mg/dL | MAE = 11.05 mg/dL
ID  2: RMSE = 17.15 mg/dL | MAE = 10.51 mg/dL
ID  3: RMSE = 10.67 mg/dL | MAE = 8.11 mg/dL
ID  4: RMSE = 9.81 mg/dL | MAE = 6.49 mg/dL
ID  5: RMSE = 12.82 mg/dL | MAE = 9.92 mg/dL
ID  6: RMSE = 11.27 mg/dL | MAE = 7.35 mg/dL
ID  7: RMSE = 11.93 mg/dL | MAE = 7.60 mg/dL
ID  8: RMSE = 9.57 mg/dL | MAE = 6.76 mg/dL
ID  9: RMSE = 14.89 mg/dL | MAE = 9.79 mg/dL
ID 10: RMSE = 11.62 mg/dL | MAE = 7.21 mg/dL
ID 11: RMSE = 15.25 mg/dL | MAE = 9.69 mg/dL
ID 12: RMSE = 12.11 mg/dL | MAE = 9.04 mg/dL
ID 13: RMSE = 14.52 mg/dL | MAE = 10.96 mg/dL
ID 14: RMSE = 12.59 mg/dL | MAE = 9.17 mg/dL
ID 15: RMSE = 14.10 mg/dL | MAE = 10.37 mg/dL
ID 16: RMSE = 12.68 mg/dL | MAE = 8.27 mg/dL
ID 17: RMSE = 16.01 mg/dL | MAE = 11.48 mg/dL
ID 18: RMSE = 16.26 mg/dL | MAE = 12.50 mg/dL
ID 19: RMSE = 12.05 mg/dL | MAE = 8.60 mg/dL
ID 20: RMSE = 17.96 mg/dL | MAE = 12.19 mg/dL
ID 21: RMSE = 10.90 mg/dL | MAE = 5.73 mg/dL
ID 22: RMSE = 10.29 mg/dL | MAE = 6.90 mg/dL
ID 23: RMSE = 16.13 mg/dL | MAE = 9.44 mg/dL
ID 24: RMSE = 15.46 mg/dL | MAE = 11.53 mg/dL
ID 25: RMSE = 11.60 mg/dL | MAE = 7.45 mg/dL
ID 26: RMSE = 6.40 mg/dL | MAE = 4.42 mg/dL
ID 27: RMSE = 16.05 mg/dL | MAE = 11.26 mg/dL
ID 28: RMSE = 10.91 mg/dL | MAE = 7.42 mg/dL
ID 29: RMSE = 14.21 mg/dL | MAE = 8.58 mg/dL
ID 30: RMSE = 12.04 mg/dL | MAE = 7.57 mg/dL
ID 31: RMSE = 10.66 mg/dL | MAE = 7.32 mg/dL
ID 32: RMSE = 14.41 mg/dL | MAE = 9.25 mg/dL
ID 33: RMSE = 14.76 mg/dL | MAE = 10.29 mg/dL
ID 34: RMSE = 14.96 mg/dL | MAE = 7.32 mg/dL
ID 35: RMSE = 6.96 mg/dL | MAE = 4.71 mg/dL
ID 36: RMSE = 10.87 mg/dL | MAE = 7.41 mg/dL
ID 37: RMSE = 12.60 mg/dL | MAE = 7.47 mg/dL
ID 38: RMSE = 10.14 mg/dL | MAE = 7.75 mg/dL
ID 39: RMSE = 13.51 mg/dL | MAE = 8.17 mg/dL
ID 40: RMSE = 12.15 mg/dL | MAE = 7.27 mg/dL
ID 41: RMSE = 11.05 mg/dL | MAE = 7.12 mg/dL
ID 42: RMSE = 13.16 mg/dL | MAE = 9.02 mg/dL
ID 43: RMSE = 10.63 mg/dL | MAE = 7.11 mg/dL
ID 44: RMSE = 7.58 mg/dL | MAE = 5.21 mg/dL
ID 45: RMSE = 10.84 mg/dL | MAE = 8.48 mg/dL
ID 46: RMSE = 11.43 mg/dL | MAE = 7.97 mg/dL
ID 47: RMSE = 9.61 mg/dL | MAE = 6.12 mg/dL
ID 48: RMSE = 9.39 mg/dL | MAE = 6.44 mg/dL
ID 49: RMSE = 15.50 mg/dL | MAE = 8.69 mg/dL
ID 50: RMSE = 13.63 mg/dL | MAE = 9.28 mg/dL
ID 51: RMSE = 9.50 mg/dL | MAE = 6.03 mg/dL
ID 52: RMSE = 13.59 mg/dL | MAE = 9.99 mg/dL
ID 53: RMSE = 11.74 mg/dL | MAE = 8.22 mg/dL
ID 54: RMSE = 12.62 mg/dL | MAE = 8.41 mg/dL
ID 55: RMSE = 10.28 mg/dL | MAE = 6.27 mg/dL
ID 56: RMSE = 13.03 mg/dL | MAE = 9.65 mg/dL


The min RMSE is for ID 26 with a value of: 6.40
The max RMSE is for ID 1 with a value of: 18.19
The min MAE  is for ID 26 with a value of: 4.42
The max MAE  is for ID 18 with a value of: 12.50


all_preds3 = []
all_truths3 = []
seq_length_in = 24 # Training on 120 mins of data (2 hours)
seq_length_out = 6 # 30 mins (~5 min interval each interation)
results3 = {}

for key, subset in normalized_data.items():
    print(f"\nRunning model for: {key}")
    
    try:
        preds, truths = RNN_indv(subset, seq_length_in, seq_length_out)
        all_preds3.append(preds)
        all_truths3.append(truths)
        results3[key] = (preds, truths)
    except Exception as e:
        print(f"Error for {key}: {e}")
        results3[key] = None

Running model for: id_0_data
Execution time: 26.958310842514038 seconds
46/46 ━━━━━━━━━━━━━━━━━━━━ 1s 8ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 5ms/step 
Execution time: 0.7195909023284912 seconds
Execution time: 0.0009069442749023438 seconds
Execution time: 0.00018310546875 seconds


Overall Train RMSE: 8.919583351218614 mg/DL
Overall Test RMSE: 7.444172863969002 mg/DL


Train RMSE at t+1: 3.85 mg/DL
Train RMSE at t+2: 5.59 mg/DL
Train RMSE at t+3: 7.43 mg/DL
Train RMSE at t+4: 9.35 mg/DL
Train RMSE at t+5: 11.26 mg/DL
Train RMSE at t+6: 12.73 mg/DL


Test RMSE at t+1: 3.31 mg/DL
Test RMSE at t+2: 4.44 mg/DL
Test RMSE at t+3: 6.14 mg/DL
Test RMSE at t+4: 7.87 mg/DL
Test RMSE at t+5: 9.20 mg/DL
Test RMSE at t+6: 10.84 mg/DL

Running model for: id_1_data
Execution time: 27.108065128326416 seconds
45/45 ━━━━━━━━━━━━━━━━━━━━ 1s 8ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 5ms/step 
Execution time: 0.7049117088317871 seconds
Execution time: 0.00074005126953125 seconds
Execution time: 0.00015211105346679688 seconds


Overall Train RMSE: 12.257854159181228 mg/DL
Overall Test RMSE: 17.39788047105271 mg/DL


Train RMSE at t+1: 4.74 mg/DL
Train RMSE at t+2: 7.76 mg/DL
Train RMSE at t+3: 10.34 mg/DL
Train RMSE at t+4: 12.97 mg/DL
Train RMSE at t+5: 15.45 mg/DL
Train RMSE at t+6: 17.47 mg/DL


Test RMSE at t+1: 9.23 mg/DL
Test RMSE at t+2: 13.81 mg/DL
Test RMSE at t+3: 17.25 mg/DL
Test RMSE at t+4: 18.81 mg/DL
Test RMSE at t+5: 20.26 mg/DL
Test RMSE at t+6: 21.87 mg/DL

Running model for: id_2_data
Execution time: 29.23233389854431 seconds
45/45 ━━━━━━━━━━━━━━━━━━━━ 1s 8ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 5ms/step 
Execution time: 0.7147030830383301 seconds
Execution time: 0.0006310939788818359 seconds
Execution time: 0.00015807151794433594 seconds


Overall Train RMSE: 7.857487464323328 mg/DL
Overall Test RMSE: 20.503170414860165 mg/DL


Train RMSE at t+1: 4.34 mg/DL
Train RMSE at t+2: 5.93 mg/DL
Train RMSE at t+3: 7.32 mg/DL
Train RMSE at t+4: 8.59 mg/DL
Train RMSE at t+5: 9.47 mg/DL
Train RMSE at t+6: 9.97 mg/DL


Test RMSE at t+1: 15.66 mg/DL
Test RMSE at t+2: 19.63 mg/DL
Test RMSE at t+3: 19.34 mg/DL
Test RMSE at t+4: 20.12 mg/DL
Test RMSE at t+5: 23.11 mg/DL
Test RMSE at t+6: 24.05 mg/DL

Running model for: id_3_data
Execution time: 22.450911045074463 seconds
44/44 ━━━━━━━━━━━━━━━━━━━━ 1s 8ms/step
11/11 ━━━━━━━━━━━━━━━━━━━━ 0s 5ms/step 
Execution time: 0.6914019584655762 seconds
Execution time: 0.00055694580078125 seconds
Execution time: 8.20159912109375e-05 seconds


Overall Train RMSE: 10.266028015674276 mg/DL
Overall Test RMSE: 9.575249491096828 mg/DL


Train RMSE at t+1: 5.78 mg/DL
Train RMSE at t+2: 7.98 mg/DL
Train RMSE at t+3: 9.92 mg/DL
Train RMSE at t+4: 11.15 mg/DL
Train RMSE at t+5: 12.10 mg/DL
Train RMSE at t+6: 12.89 mg/DL


Test RMSE at t+1: 5.51 mg/DL
Test RMSE at t+2: 7.52 mg/DL
Test RMSE at t+3: 9.16 mg/DL
Test RMSE at t+4: 10.22 mg/DL
Test RMSE at t+5: 11.32 mg/DL
Test RMSE at t+6: 12.11 mg/DL

Running model for: id_4_data
Execution time: 27.57988691329956 seconds
54/54 ━━━━━━━━━━━━━━━━━━━━ 1s 8ms/step
14/14 ━━━━━━━━━━━━━━━━━━━━ 0s 4ms/step 
Execution time: 0.7562010288238525 seconds
Execution time: 0.0007581710815429688 seconds
Execution time: 0.0001049041748046875 seconds


Overall Train RMSE: 11.152093985197052 mg/DL
Overall Test RMSE: 9.870794341611834 mg/DL


Train RMSE at t+1: 5.04 mg/DL
Train RMSE at t+2: 7.97 mg/DL
Train RMSE at t+3: 9.87 mg/DL
Train RMSE at t+4: 11.97 mg/DL
Train RMSE at t+5: 13.71 mg/DL
Train RMSE at t+6: 15.11 mg/DL


Test RMSE at t+1: 4.71 mg/DL
Test RMSE at t+2: 7.54 mg/DL
Test RMSE at t+3: 9.09 mg/DL
Test RMSE at t+4: 10.30 mg/DL
Test RMSE at t+5: 12.03 mg/DL
Test RMSE at t+6: 13.12 mg/DL

Running model for: id_5_data
Execution time: 26.91746211051941 seconds
44/44 ━━━━━━━━━━━━━━━━━━━━ 1s 8ms/step
11/11 ━━━━━━━━━━━━━━━━━━━━ 0s 5ms/step 
Execution time: 0.7048141956329346 seconds
Execution time: 0.0006656646728515625 seconds
Execution time: 9.131431579589844e-05 seconds


Overall Train RMSE: 14.850965015541698 mg/DL
Overall Test RMSE: 12.292225267898392 mg/DL


Train RMSE at t+1: 6.28 mg/DL
Train RMSE at t+2: 10.06 mg/DL
Train RMSE at t+3: 14.09 mg/DL
Train RMSE at t+4: 16.31 mg/DL
Train RMSE at t+5: 18.12 mg/DL
Train RMSE at t+6: 19.74 mg/DL


Test RMSE at t+1: 6.49 mg/DL
Test RMSE at t+2: 9.39 mg/DL
Test RMSE at t+3: 12.27 mg/DL
Test RMSE at t+4: 13.60 mg/DL
Test RMSE at t+5: 14.36 mg/DL
Test RMSE at t+6: 15.31 mg/DL

Running model for: id_6_data
Execution time: 25.571537017822266 seconds
46/46 ━━━━━━━━━━━━━━━━━━━━ 1s 8ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 5ms/step 
Execution time: 0.6953840255737305 seconds
Execution time: 0.0006771087646484375 seconds
Execution time: 0.0001850128173828125 seconds


Overall Train RMSE: 9.477779449237024 mg/DL
Overall Test RMSE: 10.905233012955387 mg/DL


Train RMSE at t+1: 3.73 mg/DL
Train RMSE at t+2: 5.89 mg/DL
Train RMSE at t+3: 7.99 mg/DL
Train RMSE at t+4: 10.05 mg/DL
Train RMSE at t+5: 11.97 mg/DL
Train RMSE at t+6: 13.50 mg/DL


Test RMSE at t+1: 5.32 mg/DL
Test RMSE at t+2: 7.83 mg/DL
Test RMSE at t+3: 10.04 mg/DL
Test RMSE at t+4: 12.01 mg/DL
Test RMSE at t+5: 13.44 mg/DL
Test RMSE at t+6: 14.09 mg/DL

Running model for: id_7_data
Execution time: 23.299540042877197 seconds
45/45 ━━━━━━━━━━━━━━━━━━━━ 1s 8ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 5ms/step 
Execution time: 0.6623082160949707 seconds
Execution time: 0.0006139278411865234 seconds
Execution time: 0.00016880035400390625 seconds


Overall Train RMSE: 13.97822128708026 mg/DL
Overall Test RMSE: 13.038182757751098 mg/DL


Train RMSE at t+1: 9.38 mg/DL
Train RMSE at t+2: 11.71 mg/DL
Train RMSE at t+3: 13.48 mg/DL
Train RMSE at t+4: 14.61 mg/DL
Train RMSE at t+5: 16.02 mg/DL
Train RMSE at t+6: 17.19 mg/DL


Test RMSE at t+1: 10.10 mg/DL
Test RMSE at t+2: 11.26 mg/DL
Test RMSE at t+3: 12.68 mg/DL
Test RMSE at t+4: 13.05 mg/DL
Test RMSE at t+5: 14.57 mg/DL
Test RMSE at t+6: 15.74 mg/DL

Running model for: id_8_data
Execution time: 22.71846914291382 seconds
39/39 ━━━━━━━━━━━━━━━━━━━━ 0s 8ms/step
10/10 ━━━━━━━━━━━━━━━━━━━━ 0s 5ms/step 
Execution time: 0.601362943649292 seconds
Execution time: 0.0005741119384765625 seconds
Execution time: 0.00013113021850585938 seconds


Overall Train RMSE: 10.911206806872185 mg/DL
Overall Test RMSE: 9.604058924932968 mg/DL


Train RMSE at t+1: 4.35 mg/DL
Train RMSE at t+2: 7.13 mg/DL
Train RMSE at t+3: 9.50 mg/DL
Train RMSE at t+4: 11.88 mg/DL
Train RMSE at t+5: 13.68 mg/DL
Train RMSE at t+6: 15.03 mg/DL


Test RMSE at t+1: 4.03 mg/DL
Test RMSE at t+2: 6.46 mg/DL
Test RMSE at t+3: 8.72 mg/DL
Test RMSE at t+4: 10.79 mg/DL
Test RMSE at t+5: 11.93 mg/DL
Test RMSE at t+6: 12.67 mg/DL

Running model for: id_9_data
Execution time: 24.64575695991516 seconds
47/47 ━━━━━━━━━━━━━━━━━━━━ 1s 9ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 5ms/step 
Execution time: 0.732048749923706 seconds
Execution time: 0.0006182193756103516 seconds
Execution time: 0.00016880035400390625 seconds


Overall Train RMSE: 11.694506565630991 mg/DL
Overall Test RMSE: 14.293989351512664 mg/DL


Train RMSE at t+1: 5.13 mg/DL
Train RMSE at t+2: 7.85 mg/DL
Train RMSE at t+3: 10.27 mg/DL
Train RMSE at t+4: 12.57 mg/DL
Train RMSE at t+5: 14.37 mg/DL
Train RMSE at t+6: 16.20 mg/DL


Test RMSE at t+1: 7.27 mg/DL
Test RMSE at t+2: 10.47 mg/DL
Test RMSE at t+3: 13.21 mg/DL
Test RMSE at t+4: 15.20 mg/DL
Test RMSE at t+5: 17.03 mg/DL
Test RMSE at t+6: 19.18 mg/DL


# Concatenate all predictions and truths
all_preds_concat3 = np.concatenate(all_preds3, axis=0).flatten()
all_truths_concat3 = np.concatenate(all_truths3, axis=0).flatten()

# Compute overall RMSE and MAE
combined_rmse3 = np.sqrt(mean_squared_error(all_truths_concat3, all_preds_concat3))
combined_mae3 = mean_absolute_error(all_truths_concat3, all_preds_concat3)

print(f"\nCombined RMSE across all IDs: {combined_rmse3:.2f} mg/dL")
print(f"Combined MAE  across all IDs: {combined_mae3:.2f} mg/dL\n")

# Lists to store RMSE and MAE per individual
rmse_per_individual = []
mae_per_individual = []

# Compute per-ID RMSE and MAE
for pred_array, true_array in zip(all_preds3, all_truths3):
    pred_flat = pred_array.flatten()
    true_flat = true_array.flatten()
    
    rmse = np.sqrt(mean_squared_error(true_flat, pred_flat))
    mae = mean_absolute_error(true_flat, pred_flat)
    
    rmse_per_individual.append(rmse)
    mae_per_individual.append(mae)

# Print all 56 RMSEs and MAEs
for i, (rmse, mae) in enumerate(zip(rmse_per_individual, mae_per_individual)):
    print(f"ID {i:2d}: RMSE = {rmse:.2f} mg/dL | MAE = {mae:.2f} mg/dL")

# Summary: min/max RMSE and MAE
print('\n')
print(f'The min RMSE is for ID {np.argmin(rmse_per_individual)} with a value of: {min(rmse_per_individual):.2f}')
print(f'The max RMSE is for ID {np.argmax(rmse_per_individual)} with a value of: {max(rmse_per_individual):.2f}')
print(f'The min MAE  is for ID {np.argmin(mae_per_individual)} with a value of: {min(mae_per_individual):.2f}')
print(f'The max MAE  is for ID {np.argmax(mae_per_individual)} with a value of: {max(mae_per_individual):.2f}')

Combined RMSE across all IDs: 12.84 mg/dL
Combined MAE  across all IDs: 8.44 mg/dL

ID  0: RMSE = 7.44 mg/dL | MAE = 5.54 mg/dL
ID  1: RMSE = 17.40 mg/dL | MAE = 10.71 mg/dL
ID  2: RMSE = 20.50 mg/dL | MAE = 14.35 mg/dL
ID  3: RMSE = 9.58 mg/dL | MAE = 7.28 mg/dL
ID  4: RMSE = 9.87 mg/dL | MAE = 6.68 mg/dL
ID  5: RMSE = 12.29 mg/dL | MAE = 9.19 mg/dL
ID  6: RMSE = 10.91 mg/dL | MAE = 6.70 mg/dL
ID  7: RMSE = 13.04 mg/dL | MAE = 9.52 mg/dL
ID  8: RMSE = 9.60 mg/dL | MAE = 6.83 mg/dL
ID  9: RMSE = 14.29 mg/dL | MAE = 9.12 mg/dL
ID 10: RMSE = 11.26 mg/dL | MAE = 8.26 mg/dL
ID 11: RMSE = 15.91 mg/dL | MAE = 10.70 mg/dL
ID 12: RMSE = 11.77 mg/dL | MAE = 8.85 mg/dL
ID 13: RMSE = 15.97 mg/dL | MAE = 12.63 mg/dL
ID 14: RMSE = 12.29 mg/dL | MAE = 8.98 mg/dL
ID 15: RMSE = 13.06 mg/dL | MAE = 9.35 mg/dL
ID 16: RMSE = 12.60 mg/dL | MAE = 8.36 mg/dL
ID 17: RMSE = 15.27 mg/dL | MAE = 10.92 mg/dL
ID 18: RMSE = 14.38 mg/dL | MAE = 9.92 mg/dL
ID 19: RMSE = 12.38 mg/dL | MAE = 8.60 mg/dL
ID 20: RMSE = 23.04 mg/dL | MAE = 12.01 mg/dL
ID 21: RMSE = 11.41 mg/dL | MAE = 7.54 mg/dL
ID 22: RMSE = 13.46 mg/dL | MAE = 9.82 mg/dL
ID 23: RMSE = 17.06 mg/dL | MAE = 9.77 mg/dL
ID 24: RMSE = 13.42 mg/dL | MAE = 8.63 mg/dL
ID 25: RMSE = 11.15 mg/dL | MAE = 7.74 mg/dL
ID 26: RMSE = 6.72 mg/dL | MAE = 5.26 mg/dL
ID 27: RMSE = 12.93 mg/dL | MAE = 7.72 mg/dL
ID 28: RMSE = 10.39 mg/dL | MAE = 6.69 mg/dL
ID 29: RMSE = 14.12 mg/dL | MAE = 8.60 mg/dL
ID 30: RMSE = 12.13 mg/dL | MAE = 7.98 mg/dL
ID 31: RMSE = 12.58 mg/dL | MAE = 9.50 mg/dL
ID 32: RMSE = 12.62 mg/dL | MAE = 7.82 mg/dL
ID 33: RMSE = 15.02 mg/dL | MAE = 10.57 mg/dL
ID 34: RMSE = 14.68 mg/dL | MAE = 7.51 mg/dL
ID 35: RMSE = 7.31 mg/dL | MAE = 4.90 mg/dL
ID 36: RMSE = 10.65 mg/dL | MAE = 7.09 mg/dL
ID 37: RMSE = 12.34 mg/dL | MAE = 7.31 mg/dL
ID 38: RMSE = 10.28 mg/dL | MAE = 7.90 mg/dL
ID 39: RMSE = 13.87 mg/dL | MAE = 8.35 mg/dL
ID 40: RMSE = 10.99 mg/dL | MAE = 7.85 mg/dL
ID 41: RMSE = 12.98 mg/dL | MAE = 9.42 mg/dL
ID 42: RMSE = 12.92 mg/dL | MAE = 9.35 mg/dL
ID 43: RMSE = 9.62 mg/dL | MAE = 6.55 mg/dL
ID 44: RMSE = 7.32 mg/dL | MAE = 5.18 mg/dL
ID 45: RMSE = 10.86 mg/dL | MAE = 8.53 mg/dL
ID 46: RMSE = 11.52 mg/dL | MAE = 8.10 mg/dL
ID 47: RMSE = 9.62 mg/dL | MAE = 6.01 mg/dL
ID 48: RMSE = 9.21 mg/dL | MAE = 5.96 mg/dL
ID 49: RMSE = 14.92 mg/dL | MAE = 8.30 mg/dL
ID 50: RMSE = 13.37 mg/dL | MAE = 9.83 mg/dL
ID 51: RMSE = 9.68 mg/dL | MAE = 6.27 mg/dL
ID 52: RMSE = 13.13 mg/dL | MAE = 8.88 mg/dL
ID 53: RMSE = 15.20 mg/dL | MAE = 11.91 mg/dL
ID 54: RMSE = 12.58 mg/dL | MAE = 8.29 mg/dL
ID 55: RMSE = 9.89 mg/dL | MAE = 5.86 mg/dL
ID 56: RMSE = 12.13 mg/dL | MAE = 8.46 mg/dL


The min RMSE is for ID 26 with a value of: 6.72
The max RMSE is for ID 20 with a value of: 23.04
The min MAE  is for ID 35 with a value of: 4.90
The max MAE  is for ID 2 with a value of: 14.35


all_preds4 = []
all_truths4 = []
seq_length_in = 12
seq_length_out = 6
results4 = {}

for key, subset in normalized_data.items():
    print(f"\nRunning model for: {key}")
    
    try:
        preds, truths = RNN_indv(subset, seq_length_in, seq_length_out)
        all_preds4.append(preds)
        all_truths4.append(truths)
        results4[key] = (preds, truths)
    except Exception as e:
        print(f"Error for {key}: {e}")
        results4[key] = None

Running model for: id_0_data
Execution time: 12.6473970413208 seconds
46/46 ━━━━━━━━━━━━━━━━━━━━ 0s 7ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.5881669521331787 seconds
Execution time: 0.0006020069122314453 seconds
Execution time: 0.0001380443572998047 seconds


Overall Train RMSE: 8.913581060702924 mg/DL
Overall Test RMSE: 7.693591106708841 mg/DL


Train RMSE at t+1: 4.13 mg/DL
Train RMSE at t+2: 5.23 mg/DL
Train RMSE at t+3: 7.18 mg/DL
Train RMSE at t+4: 9.60 mg/DL
Train RMSE at t+5: 11.36 mg/DL
Train RMSE at t+6: 12.63 mg/DL


Test RMSE at t+1: 3.43 mg/DL
Test RMSE at t+2: 4.17 mg/DL
Test RMSE at t+3: 5.99 mg/DL
Test RMSE at t+4: 8.38 mg/DL
Test RMSE at t+5: 9.84 mg/DL
Test RMSE at t+6: 11.09 mg/DL

Running model for: id_1_data
Execution time: 15.114711999893188 seconds
45/45 ━━━━━━━━━━━━━━━━━━━━ 0s 6ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.5600149631500244 seconds
Execution time: 0.0006470680236816406 seconds
Execution time: 9.179115295410156e-05 seconds


Overall Train RMSE: 12.512771393246899 mg/DL
Overall Test RMSE: 17.539543637887032 mg/DL


Train RMSE at t+1: 4.88 mg/DL
Train RMSE at t+2: 8.24 mg/DL
Train RMSE at t+3: 10.67 mg/DL
Train RMSE at t+4: 13.30 mg/DL
Train RMSE at t+5: 15.62 mg/DL
Train RMSE at t+6: 17.69 mg/DL


Test RMSE at t+1: 9.57 mg/DL
Test RMSE at t+2: 14.18 mg/DL
Test RMSE at t+3: 16.90 mg/DL
Test RMSE at t+4: 18.84 mg/DL
Test RMSE at t+5: 20.77 mg/DL
Test RMSE at t+6: 21.93 mg/DL

Running model for: id_2_data
Execution time: 14.90549898147583 seconds
46/46 ━━━━━━━━━━━━━━━━━━━━ 0s 6ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.5539107322692871 seconds
Execution time: 0.0006749629974365234 seconds
Execution time: 9.107589721679688e-05 seconds


Overall Train RMSE: 8.076921756708067 mg/DL
Overall Test RMSE: 17.04248139723138 mg/DL


Train RMSE at t+1: 3.95 mg/DL
Train RMSE at t+2: 5.86 mg/DL
Train RMSE at t+3: 7.41 mg/DL
Train RMSE at t+4: 8.79 mg/DL
Train RMSE at t+5: 9.81 mg/DL
Train RMSE at t+6: 10.63 mg/DL


Test RMSE at t+1: 9.87 mg/DL
Test RMSE at t+2: 13.10 mg/DL
Test RMSE at t+3: 16.13 mg/DL
Test RMSE at t+4: 18.63 mg/DL
Test RMSE at t+5: 20.10 mg/DL
Test RMSE at t+6: 21.51 mg/DL

Running model for: id_3_data
Execution time: 15.196372032165527 seconds
45/45 ━━━━━━━━━━━━━━━━━━━━ 0s 6ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.5587503910064697 seconds
Execution time: 0.0006451606750488281 seconds
Execution time: 8.702278137207031e-05 seconds


Overall Train RMSE: 10.120584908061227 mg/DL
Overall Test RMSE: 9.381126168366832 mg/DL


Train RMSE at t+1: 5.65 mg/DL
Train RMSE at t+2: 7.65 mg/DL
Train RMSE at t+3: 9.47 mg/DL
Train RMSE at t+4: 10.97 mg/DL
Train RMSE at t+5: 12.14 mg/DL
Train RMSE at t+6: 12.92 mg/DL


Test RMSE at t+1: 5.72 mg/DL
Test RMSE at t+2: 7.42 mg/DL
Test RMSE at t+3: 8.75 mg/DL
Test RMSE at t+4: 9.97 mg/DL
Test RMSE at t+5: 11.14 mg/DL
Test RMSE at t+6: 11.84 mg/DL

Running model for: id_4_data
Execution time: 15.290265083312988 seconds
55/55 ━━━━━━━━━━━━━━━━━━━━ 0s 5ms/step
14/14 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.5812041759490967 seconds
Execution time: 0.0007219314575195312 seconds
Execution time: 0.0002071857452392578 seconds


Overall Train RMSE: 11.878325681076188 mg/DL
Overall Test RMSE: 9.663083251073235 mg/DL


Train RMSE at t+1: 5.19 mg/DL
Train RMSE at t+2: 7.58 mg/DL
Train RMSE at t+3: 11.41 mg/DL
Train RMSE at t+4: 12.86 mg/DL
Train RMSE at t+5: 14.27 mg/DL
Train RMSE at t+6: 16.22 mg/DL


Test RMSE at t+1: 4.71 mg/DL
Test RMSE at t+2: 6.81 mg/DL
Test RMSE at t+3: 8.82 mg/DL
Test RMSE at t+4: 10.32 mg/DL
Test RMSE at t+5: 11.81 mg/DL
Test RMSE at t+6: 12.96 mg/DL

Running model for: id_5_data
Execution time: 15.10288691520691 seconds
45/45 ━━━━━━━━━━━━━━━━━━━━ 0s 6ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.5635890960693359 seconds
Execution time: 0.0005970001220703125 seconds
Execution time: 8.487701416015625e-05 seconds


Overall Train RMSE: 14.922526500639796 mg/DL
Overall Test RMSE: 11.700163559626706 mg/DL


Train RMSE at t+1: 6.99 mg/DL
Train RMSE at t+2: 10.46 mg/DL
Train RMSE at t+3: 13.76 mg/DL
Train RMSE at t+4: 16.07 mg/DL
Train RMSE at t+5: 18.28 mg/DL
Train RMSE at t+6: 19.89 mg/DL


Test RMSE at t+1: 6.67 mg/DL
Test RMSE at t+2: 9.23 mg/DL
Test RMSE at t+3: 11.35 mg/DL
Test RMSE at t+4: 12.75 mg/DL
Test RMSE at t+5: 13.66 mg/DL
Test RMSE at t+6: 14.62 mg/DL

Running model for: id_6_data
Execution time: 15.138405799865723 seconds
47/47 ━━━━━━━━━━━━━━━━━━━━ 0s 6ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.5589210987091064 seconds
Execution time: 0.0005440711975097656 seconds
Execution time: 0.00016927719116210938 seconds


Overall Train RMSE: 9.835546611439684 mg/DL
Overall Test RMSE: 10.583125405779654 mg/DL


Train RMSE at t+1: 4.75 mg/DL
Train RMSE at t+2: 6.44 mg/DL
Train RMSE at t+3: 8.47 mg/DL
Train RMSE at t+4: 10.42 mg/DL
Train RMSE at t+5: 12.13 mg/DL
Train RMSE at t+6: 13.74 mg/DL


Test RMSE at t+1: 5.90 mg/DL
Test RMSE at t+2: 7.74 mg/DL
Test RMSE at t+3: 9.70 mg/DL
Test RMSE at t+4: 11.62 mg/DL
Test RMSE at t+5: 12.70 mg/DL
Test RMSE at t+6: 13.68 mg/DL

Running model for: id_7_data
Execution time: 14.726606130599976 seconds
46/46 ━━━━━━━━━━━━━━━━━━━━ 0s 6ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.5585329532623291 seconds
Execution time: 0.0005786418914794922 seconds
Execution time: 8.606910705566406e-05 seconds


Overall Train RMSE: 14.17623140621529 mg/DL
Overall Test RMSE: 10.649737566020159 mg/DL


Train RMSE at t+1: 9.44 mg/DL
Train RMSE at t+2: 10.13 mg/DL
Train RMSE at t+3: 13.10 mg/DL
Train RMSE at t+4: 15.40 mg/DL
Train RMSE at t+5: 16.86 mg/DL
Train RMSE at t+6: 17.92 mg/DL


Test RMSE at t+1: 8.49 mg/DL
Test RMSE at t+2: 7.02 mg/DL
Test RMSE at t+3: 9.42 mg/DL
Test RMSE at t+4: 11.32 mg/DL
Test RMSE at t+5: 12.68 mg/DL
Test RMSE at t+6: 13.46 mg/DL

Running model for: id_8_data
Execution time: 15.08035683631897 seconds
40/40 ━━━━━━━━━━━━━━━━━━━━ 0s 7ms/step
10/10 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.5463550090789795 seconds
Execution time: 0.0006928443908691406 seconds
Execution time: 7.987022399902344e-05 seconds


Overall Train RMSE: 11.865006543472456 mg/DL
Overall Test RMSE: 9.856293594572396 mg/DL


Train RMSE at t+1: 4.93 mg/DL
Train RMSE at t+2: 7.94 mg/DL
Train RMSE at t+3: 10.95 mg/DL
Train RMSE at t+4: 12.98 mg/DL
Train RMSE at t+5: 14.59 mg/DL
Train RMSE at t+6: 16.00 mg/DL


Test RMSE at t+1: 4.31 mg/DL
Test RMSE at t+2: 6.95 mg/DL
Test RMSE at t+3: 9.25 mg/DL
Test RMSE at t+4: 10.96 mg/DL
Test RMSE at t+5: 11.97 mg/DL
Test RMSE at t+6: 12.92 mg/DL

Running model for: id_9_data
Execution time: 15.234620809555054 seconds
47/47 ━━━━━━━━━━━━━━━━━━━━ 0s 2ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 19ms/step
Execution time: 0.5659921169281006 seconds
Execution time: 0.0005371570587158203 seconds
Execution time: 0.0001392364501953125 seconds


Overall Train RMSE: 11.658591247090422 mg/DL
Overall Test RMSE: 14.283516072642591 mg/DL


Train RMSE at t+1: 5.33 mg/DL
Train RMSE at t+2: 7.77 mg/DL
Train RMSE at t+3: 10.31 mg/DL
Train RMSE at t+4: 12.43 mg/DL
Train RMSE at t+5: 14.45 mg/DL
Train RMSE at t+6: 16.04 mg/DL


Test RMSE at t+1: 7.70 mg/DL
Test RMSE at t+2: 10.72 mg/DL
Test RMSE at t+3: 13.24 mg/DL
Test RMSE at t+4: 15.24 mg/DL
Test RMSE at t+5: 16.95 mg/DL
Test RMSE at t+6: 18.85 mg/DL


# Concatenate all predictions and truths
all_preds_concat4 = np.concatenate(all_preds4, axis=0).flatten()
all_truths_concat4 = np.concatenate(all_truths4, axis=0).flatten()

# Compute overall RMSE and MAE
combined_rmse4 = np.sqrt(mean_squared_error(all_truths_concat4, all_preds_concat4))
combined_mae4 = mean_absolute_error(all_truths_concat4, all_preds_concat4)

print(f"\nCombined RMSE across all IDs: {combined_rmse4:.2f} mg/dL")
print(f"Combined MAE  across all IDs: {combined_mae4:.2f} mg/dL\n")

# Lists to store RMSE and MAE per individual
rmse_per_individual = []
mae_per_individual = []

# Compute per-ID RMSE and MAE
for pred_array, true_array in zip(all_preds4, all_truths4):
    pred_flat = pred_array.flatten()
    true_flat = true_array.flatten()
    
    rmse = np.sqrt(mean_squared_error(true_flat, pred_flat))
    mae = mean_absolute_error(true_flat, pred_flat)
    
    rmse_per_individual.append(rmse)
    mae_per_individual.append(mae)

# Print all 56 RMSEs and MAEs
for i, (rmse, mae) in enumerate(zip(rmse_per_individual, mae_per_individual)):
    print(f"ID {i:2d}: RMSE = {rmse:.2f} mg/dL | MAE = {mae:.2f} mg/dL")

# Summary: min/max RMSE and MAE
print('\n')
print(f'The min RMSE is for ID {np.argmin(rmse_per_individual)} with a value of: {min(rmse_per_individual):.2f}')
print(f'The max RMSE is for ID {np.argmax(rmse_per_individual)} with a value of: {max(rmse_per_individual):.2f}')
print(f'The min MAE  is for ID {np.argmin(mae_per_individual)} with a value of: {min(mae_per_individual):.2f}')
print(f'The max MAE  is for ID {np.argmax(mae_per_individual)} with a value of: {max(mae_per_individual):.2f}')

Combined RMSE across all IDs: 12.26 mg/dL
Combined MAE  across all IDs: 8.01 mg/dL

ID  0: RMSE = 7.69 mg/dL | MAE = 5.37 mg/dL
ID  1: RMSE = 17.54 mg/dL | MAE = 11.07 mg/dL
ID  2: RMSE = 17.04 mg/dL | MAE = 10.27 mg/dL
ID  3: RMSE = 9.38 mg/dL | MAE = 7.10 mg/dL
ID  4: RMSE = 9.66 mg/dL | MAE = 6.51 mg/dL
ID  5: RMSE = 11.70 mg/dL | MAE = 8.43 mg/dL
ID  6: RMSE = 10.58 mg/dL | MAE = 6.83 mg/dL
ID  7: RMSE = 10.65 mg/dL | MAE = 8.22 mg/dL
ID  8: RMSE = 9.86 mg/dL | MAE = 7.07 mg/dL
ID  9: RMSE = 14.28 mg/dL | MAE = 8.79 mg/dL
ID 10: RMSE = 10.61 mg/dL | MAE = 7.36 mg/dL
ID 11: RMSE = 15.66 mg/dL | MAE = 9.88 mg/dL
ID 12: RMSE = 11.95 mg/dL | MAE = 9.13 mg/dL
ID 13: RMSE = 14.62 mg/dL | MAE = 11.11 mg/dL
ID 14: RMSE = 12.25 mg/dL | MAE = 8.70 mg/dL
ID 15: RMSE = 13.79 mg/dL | MAE = 10.05 mg/dL
ID 16: RMSE = 12.49 mg/dL | MAE = 8.29 mg/dL
ID 17: RMSE = 15.26 mg/dL | MAE = 10.92 mg/dL
ID 18: RMSE = 14.49 mg/dL | MAE = 9.91 mg/dL
ID 19: RMSE = 12.33 mg/dL | MAE = 8.94 mg/dL
ID 20: RMSE = 13.26 mg/dL | MAE = 7.31 mg/dL
ID 21: RMSE = 10.96 mg/dL | MAE = 7.15 mg/dL
ID 22: RMSE = 11.80 mg/dL | MAE = 7.91 mg/dL
ID 23: RMSE = 16.72 mg/dL | MAE = 9.13 mg/dL
ID 24: RMSE = 13.68 mg/dL | MAE = 8.82 mg/dL
ID 25: RMSE = 11.21 mg/dL | MAE = 7.94 mg/dL
ID 26: RMSE = 6.43 mg/dL | MAE = 5.04 mg/dL
ID 27: RMSE = 13.41 mg/dL | MAE = 7.97 mg/dL
ID 28: RMSE = 10.75 mg/dL | MAE = 7.36 mg/dL
ID 29: RMSE = 13.97 mg/dL | MAE = 8.43 mg/dL
ID 30: RMSE = 11.73 mg/dL | MAE = 8.06 mg/dL
ID 31: RMSE = 10.59 mg/dL | MAE = 7.27 mg/dL
ID 32: RMSE = 13.00 mg/dL | MAE = 7.72 mg/dL
ID 33: RMSE = 14.59 mg/dL | MAE = 9.87 mg/dL
ID 34: RMSE = 14.23 mg/dL | MAE = 7.36 mg/dL
ID 35: RMSE = 7.07 mg/dL | MAE = 4.67 mg/dL
ID 36: RMSE = 11.40 mg/dL | MAE = 7.44 mg/dL
ID 37: RMSE = 11.75 mg/dL | MAE = 6.83 mg/dL
ID 38: RMSE = 8.80 mg/dL | MAE = 6.33 mg/dL
ID 39: RMSE = 13.68 mg/dL | MAE = 7.94 mg/dL
ID 40: RMSE = 9.96 mg/dL | MAE = 6.31 mg/dL
ID 41: RMSE = 12.20 mg/dL | MAE = 7.39 mg/dL
ID 42: RMSE = 12.23 mg/dL | MAE = 8.09 mg/dL
ID 43: RMSE = 9.73 mg/dL | MAE = 6.91 mg/dL
ID 44: RMSE = 7.49 mg/dL | MAE = 5.09 mg/dL
ID 45: RMSE = 12.97 mg/dL | MAE = 10.26 mg/dL
ID 46: RMSE = 11.29 mg/dL | MAE = 7.73 mg/dL
ID 47: RMSE = 9.12 mg/dL | MAE = 5.57 mg/dL
ID 48: RMSE = 9.44 mg/dL | MAE = 6.40 mg/dL
ID 49: RMSE = 15.43 mg/dL | MAE = 9.94 mg/dL
ID 50: RMSE = 13.19 mg/dL | MAE = 8.73 mg/dL
ID 51: RMSE = 9.75 mg/dL | MAE = 6.18 mg/dL
ID 52: RMSE = 13.00 mg/dL | MAE = 9.56 mg/dL
ID 53: RMSE = 12.93 mg/dL | MAE = 8.97 mg/dL
ID 54: RMSE = 12.91 mg/dL | MAE = 9.37 mg/dL
ID 55: RMSE = 9.69 mg/dL | MAE = 6.12 mg/dL
ID 56: RMSE = 11.83 mg/dL | MAE = 8.14 mg/dL


The min RMSE is for ID 26 with a value of: 6.43
The max RMSE is for ID 1 with a value of: 17.54
The min MAE  is for ID 35 with a value of: 4.67
The max MAE  is for ID 13 with a value of: 11.11


all_preds5 = []
all_truths5 = []
seq_length_in = 6
seq_length_out = 6
results5 = {}

for key, subset in normalized_data.items():
    print(f"\nRunning model for: {key}")
    
    try:
        preds, truths = RNN_indv(subset, seq_length_in, seq_length_out)
        all_preds5.append(preds)
        all_truths5.append(truths)
        results5[key] = (preds, truths)
    except Exception as e:
        print(f"Error for {key}: {e}")
        results5[key] = None

Running model for: id_0_data
Execution time: 16.3315691947937 seconds
46/46 ━━━━━━━━━━━━━━━━━━━━ 1s 9ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.9667301177978516 seconds
Execution time: 0.003509044647216797 seconds
Execution time: 0.0002009868621826172 seconds


Overall Train RMSE: 10.162373546024325 mg/DL
Overall Test RMSE: 9.004995142803308 mg/DL


Train RMSE at t+1: 4.75 mg/DL
Train RMSE at t+2: 6.75 mg/DL
Train RMSE at t+3: 8.73 mg/DL
Train RMSE at t+4: 10.91 mg/DL
Train RMSE at t+5: 12.62 mg/DL
Train RMSE at t+6: 14.03 mg/DL


Test RMSE at t+1: 4.39 mg/DL
Test RMSE at t+2: 6.08 mg/DL
Test RMSE at t+3: 7.61 mg/DL
Test RMSE at t+4: 9.71 mg/DL
Test RMSE at t+5: 11.15 mg/DL
Test RMSE at t+6: 12.40 mg/DL

Running model for: id_1_data
Execution time: 12.848817825317383 seconds
45/45 ━━━━━━━━━━━━━━━━━━━━ 1s 8ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.7461161613464355 seconds
Execution time: 0.0007748603820800781 seconds
Execution time: 0.00017213821411132812 seconds


Overall Train RMSE: 14.45717534448805 mg/DL
Overall Test RMSE: 18.553859553929787 mg/DL


Train RMSE at t+1: 8.33 mg/DL
Train RMSE at t+2: 10.61 mg/DL
Train RMSE at t+3: 13.11 mg/DL
Train RMSE at t+4: 15.15 mg/DL
Train RMSE at t+5: 17.38 mg/DL
Train RMSE at t+6: 19.19 mg/DL


Test RMSE at t+1: 12.32 mg/DL
Test RMSE at t+2: 15.63 mg/DL
Test RMSE at t+3: 18.09 mg/DL
Test RMSE at t+4: 19.73 mg/DL
Test RMSE at t+5: 21.25 mg/DL
Test RMSE at t+6: 22.39 mg/DL

Running model for: id_2_data
Execution time: 12.239893913269043 seconds
46/46 ━━━━━━━━━━━━━━━━━━━━ 1s 7ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 2ms/step 
Execution time: 0.6825978755950928 seconds
Execution time: 0.0005660057067871094 seconds
Execution time: 0.00015687942504882812 seconds


Overall Train RMSE: 8.732570521405238 mg/DL
Overall Test RMSE: 23.152247138747644 mg/DL


Train RMSE at t+1: 5.22 mg/DL
Train RMSE at t+2: 6.91 mg/DL
Train RMSE at t+3: 8.44 mg/DL
Train RMSE at t+4: 9.46 mg/DL
Train RMSE at t+5: 10.28 mg/DL
Train RMSE at t+6: 10.77 mg/DL


Test RMSE at t+1: 17.58 mg/DL
Test RMSE at t+2: 20.00 mg/DL
Test RMSE at t+3: 23.06 mg/DL
Test RMSE at t+4: 24.64 mg/DL
Test RMSE at t+5: 25.89 mg/DL
Test RMSE at t+6: 26.42 mg/DL

Running model for: id_3_data
Execution time: 12.378146171569824 seconds
45/45 ━━━━━━━━━━━━━━━━━━━━ 0s 6ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.6128132343292236 seconds
Execution time: 0.0006279945373535156 seconds
Execution time: 0.00017905235290527344 seconds


Overall Train RMSE: 10.680856082882142 mg/DL
Overall Test RMSE: 9.417660362827649 mg/DL


Train RMSE at t+1: 5.59 mg/DL
Train RMSE at t+2: 8.00 mg/DL
Train RMSE at t+3: 9.92 mg/DL
Train RMSE at t+4: 11.71 mg/DL
Train RMSE at t+5: 12.80 mg/DL
Train RMSE at t+6: 13.78 mg/DL


Test RMSE at t+1: 5.51 mg/DL
Test RMSE at t+2: 7.56 mg/DL
Test RMSE at t+3: 8.83 mg/DL
Test RMSE at t+4: 10.07 mg/DL
Test RMSE at t+5: 11.07 mg/DL
Test RMSE at t+6: 11.94 mg/DL

Running model for: id_4_data
Execution time: 12.945731163024902 seconds
55/55 ━━━━━━━━━━━━━━━━━━━━ 0s 5ms/step
14/14 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.6345410346984863 seconds
Execution time: 0.0006730556488037109 seconds
Execution time: 0.00016498565673828125 seconds


Overall Train RMSE: 12.247248200733226 mg/DL
Overall Test RMSE: 10.37746386549608 mg/DL


Train RMSE at t+1: 6.18 mg/DL
Train RMSE at t+2: 8.67 mg/DL
Train RMSE at t+3: 11.19 mg/DL
Train RMSE at t+4: 13.01 mg/DL
Train RMSE at t+5: 15.14 mg/DL
Train RMSE at t+6: 16.22 mg/DL


Test RMSE at t+1: 5.32 mg/DL
Test RMSE at t+2: 7.79 mg/DL
Test RMSE at t+3: 9.52 mg/DL
Test RMSE at t+4: 11.45 mg/DL
Test RMSE at t+5: 12.23 mg/DL
Test RMSE at t+6: 13.64 mg/DL

Running model for: id_5_data
Execution time: 14.139282941818237 seconds
45/45 ━━━━━━━━━━━━━━━━━━━━ 0s 7ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.6419730186462402 seconds
Execution time: 0.0006330013275146484 seconds
Execution time: 0.00017404556274414062 seconds


Overall Train RMSE: 15.844616797101548 mg/DL
Overall Test RMSE: 12.289418232568279 mg/DL


Train RMSE at t+1: 8.32 mg/DL
Train RMSE at t+2: 11.98 mg/DL
Train RMSE at t+3: 14.40 mg/DL
Train RMSE at t+4: 17.52 mg/DL
Train RMSE at t+5: 19.23 mg/DL
Train RMSE at t+6: 20.23 mg/DL


Test RMSE at t+1: 7.81 mg/DL
Test RMSE at t+2: 10.42 mg/DL
Test RMSE at t+3: 11.73 mg/DL
Test RMSE at t+4: 13.60 mg/DL
Test RMSE at t+5: 14.25 mg/DL
Test RMSE at t+6: 14.53 mg/DL

Running model for: id_6_data
Execution time: 12.699170112609863 seconds
47/47 ━━━━━━━━━━━━━━━━━━━━ 0s 6ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.6079661846160889 seconds
Execution time: 0.0006499290466308594 seconds
Execution time: 0.00015997886657714844 seconds


Overall Train RMSE: 10.857564375137592 mg/DL
Overall Test RMSE: 10.696340381543386 mg/DL


Train RMSE at t+1: 5.37 mg/DL
Train RMSE at t+2: 7.49 mg/DL
Train RMSE at t+3: 9.73 mg/DL
Train RMSE at t+4: 11.67 mg/DL
Train RMSE at t+5: 13.26 mg/DL
Train RMSE at t+6: 14.69 mg/DL


Test RMSE at t+1: 6.03 mg/DL
Test RMSE at t+2: 8.31 mg/DL
Test RMSE at t+3: 10.25 mg/DL
Test RMSE at t+4: 11.70 mg/DL
Test RMSE at t+5: 12.66 mg/DL
Test RMSE at t+6: 13.37 mg/DL

Running model for: id_7_data
Execution time: 12.041678190231323 seconds
46/46 ━━━━━━━━━━━━━━━━━━━━ 1s 7ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.6337380409240723 seconds
Execution time: 0.0006451606750488281 seconds
Execution time: 9.989738464355469e-05 seconds


Overall Train RMSE: 13.61557154303043 mg/DL
Overall Test RMSE: 10.286485856677247 mg/DL


Train RMSE at t+1: 6.91 mg/DL
Train RMSE at t+2: 9.87 mg/DL
Train RMSE at t+3: 12.94 mg/DL
Train RMSE at t+4: 14.89 mg/DL
Train RMSE at t+5: 16.31 mg/DL
Train RMSE at t+6: 17.66 mg/DL


Test RMSE at t+1: 4.77 mg/DL
Test RMSE at t+2: 6.77 mg/DL
Test RMSE at t+3: 9.21 mg/DL
Test RMSE at t+4: 11.11 mg/DL
Test RMSE at t+5: 12.90 mg/DL
Test RMSE at t+6: 13.85 mg/DL

Running model for: id_8_data
Execution time: 13.344414949417114 seconds
40/40 ━━━━━━━━━━━━━━━━━━━━ 1s 9ms/step
10/10 ━━━━━━━━━━━━━━━━━━━━ 0s 4ms/step 
Execution time: 0.7971079349517822 seconds
Execution time: 0.0007562637329101562 seconds
Execution time: 0.0001010894775390625 seconds


Overall Train RMSE: 12.659784491209242 mg/DL
Overall Test RMSE: 11.445635786814297 mg/DL


Train RMSE at t+1: 7.08 mg/DL
Train RMSE at t+2: 9.43 mg/DL
Train RMSE at t+3: 11.51 mg/DL
Train RMSE at t+4: 13.78 mg/DL
Train RMSE at t+5: 15.25 mg/DL
Train RMSE at t+6: 16.37 mg/DL


Test RMSE at t+1: 6.91 mg/DL
Test RMSE at t+2: 8.87 mg/DL
Test RMSE at t+3: 10.61 mg/DL
Test RMSE at t+4: 12.60 mg/DL
Test RMSE at t+5: 13.61 mg/DL
Test RMSE at t+6: 14.24 mg/DL

Running model for: id_9_data
Execution time: 12.930803775787354 seconds
48/48 ━━━━━━━━━━━━━━━━━━━━ 1s 7ms/step
12/12 ━━━━━━━━━━━━━━━━━━━━ 0s 3ms/step 
Execution time: 0.7145299911499023 seconds
Execution time: 0.0007588863372802734 seconds
Execution time: 0.00017714500427246094 seconds


Overall Train RMSE: 13.220797532122814 mg/DL
Overall Test RMSE: 15.677843151335841 mg/DL


Train RMSE at t+1: 8.61 mg/DL
Train RMSE at t+2: 10.23 mg/DL
Train RMSE at t+3: 12.12 mg/DL
Train RMSE at t+4: 13.88 mg/DL
Train RMSE at t+5: 15.51 mg/DL
Train RMSE at t+6: 17.03 mg/DL


Test RMSE at t+1: 10.34 mg/DL
Test RMSE at t+2: 12.69 mg/DL
Test RMSE at t+3: 14.50 mg/DL
Test RMSE at t+4: 16.45 mg/DL
Test RMSE at t+5: 18.30 mg/DL
Test RMSE at t+6: 19.79 mg/DL


# Concatenate all predictions and truths
all_preds_concat1 = np.concatenate(all_preds5, axis=0).flatten()
all_truths_concat1 = np.concatenate(all_truths5, axis=0).flatten()

# Compute overall RMSE and MAE
combined_rmse5 = np.sqrt(mean_squared_error(all_truths_concat5, all_preds_concat1))
combined_mae5 = mean_absolute_error(all_truths_concat5, all_preds_concat1)

print(f"\nCombined RMSE across all IDs: {combined_rmse5:.2f} mg/dL")
print(f"Combined MAE  across all IDs: {combined_mae5:.2f} mg/dL\n")

# Lists to store RMSE and MAE per individual
rmse_per_individual = []
mae_per_individual = []

# Compute per-ID RMSE and MAE
for pred_array, true_array in zip(all_preds5, all_truths5):
    pred_flat = pred_array.flatten()
    true_flat = true_array.flatten()
    
    rmse = np.sqrt(mean_squared_error(true_flat, pred_flat))
    mae = mean_absolute_error(true_flat, pred_flat)
    
    rmse_per_individual.append(rmse)
    mae_per_individual.append(mae)

# Print all 56 RMSEs and MAEs
for i, (rmse, mae) in enumerate(zip(rmse_per_individual, mae_per_individual)):
    print(f"ID {i:2d}: RMSE = {rmse:.2f} mg/dL | MAE = {mae:.2f} mg/dL")

# Summary: min/max RMSE and MAE
print('\n')
print(f'The min RMSE is for ID {np.argmin(rmse_per_individual)} with a value of: {min(rmse_per_individual):.2f}')
print(f'The max RMSE is for ID {np.argmax(rmse_per_individual)} with a value of: {max(rmse_per_individual):.2f}')
print(f'The min MAE  is for ID {np.argmin(mae_per_individual)} with a value of: {min(mae_per_individual):.2f}')
print(f'The max MAE  is for ID {np.argmax(mae_per_individual)} with a value of: {max(mae_per_individual):.2f}')

Combined RMSE across all IDs: 12.87 mg/dL
Combined MAE  across all IDs: 8.53 mg/dL

ID  0: RMSE = 9.00 mg/dL | MAE = 7.11 mg/dL
ID  1: RMSE = 18.55 mg/dL | MAE = 12.26 mg/dL
ID  2: RMSE = 23.15 mg/dL | MAE = 17.47 mg/dL
ID  3: RMSE = 9.42 mg/dL | MAE = 7.14 mg/dL
ID  4: RMSE = 10.38 mg/dL | MAE = 7.15 mg/dL
ID  5: RMSE = 12.29 mg/dL | MAE = 9.10 mg/dL
ID  6: RMSE = 10.70 mg/dL | MAE = 6.33 mg/dL
ID  7: RMSE = 10.29 mg/dL | MAE = 7.09 mg/dL
ID  8: RMSE = 11.45 mg/dL | MAE = 8.78 mg/dL
ID  9: RMSE = 15.68 mg/dL | MAE = 10.99 mg/dL
ID 10: RMSE = 11.64 mg/dL | MAE = 8.69 mg/dL
ID 11: RMSE = 16.21 mg/dL | MAE = 10.59 mg/dL
ID 12: RMSE = 12.88 mg/dL | MAE = 9.93 mg/dL
ID 13: RMSE = 14.46 mg/dL | MAE = 10.78 mg/dL
ID 14: RMSE = 12.67 mg/dL | MAE = 9.31 mg/dL
ID 15: RMSE = 13.70 mg/dL | MAE = 10.21 mg/dL
ID 16: RMSE = 12.17 mg/dL | MAE = 7.80 mg/dL
ID 17: RMSE = 15.46 mg/dL | MAE = 11.02 mg/dL
ID 18: RMSE = 15.32 mg/dL | MAE = 10.00 mg/dL
ID 19: RMSE = 12.21 mg/dL | MAE = 8.87 mg/dL
ID 20: RMSE = 13.88 mg/dL | MAE = 8.49 mg/dL
ID 21: RMSE = 11.25 mg/dL | MAE = 5.65 mg/dL
ID 22: RMSE = 10.60 mg/dL | MAE = 7.09 mg/dL
ID 23: RMSE = 17.09 mg/dL | MAE = 9.76 mg/dL
ID 24: RMSE = 13.43 mg/dL | MAE = 8.41 mg/dL
ID 25: RMSE = 10.94 mg/dL | MAE = 7.03 mg/dL
ID 26: RMSE = 6.51 mg/dL | MAE = 4.81 mg/dL
ID 27: RMSE = 13.97 mg/dL | MAE = 8.76 mg/dL
ID 28: RMSE = 10.26 mg/dL | MAE = 6.67 mg/dL
ID 29: RMSE = 14.61 mg/dL | MAE = 8.76 mg/dL
ID 30: RMSE = 11.32 mg/dL | MAE = 7.57 mg/dL
ID 31: RMSE = 10.72 mg/dL | MAE = 7.59 mg/dL
ID 32: RMSE = 13.73 mg/dL | MAE = 8.28 mg/dL
ID 33: RMSE = 15.11 mg/dL | MAE = 10.44 mg/dL
ID 34: RMSE = 15.38 mg/dL | MAE = 8.07 mg/dL
ID 35: RMSE = 7.66 mg/dL | MAE = 5.79 mg/dL
ID 36: RMSE = 12.14 mg/dL | MAE = 8.79 mg/dL
ID 37: RMSE = 13.25 mg/dL | MAE = 7.58 mg/dL
ID 38: RMSE = 9.52 mg/dL | MAE = 6.95 mg/dL
ID 39: RMSE = 14.27 mg/dL | MAE = 8.35 mg/dL
ID 40: RMSE = 11.65 mg/dL | MAE = 6.97 mg/dL
ID 41: RMSE = 13.85 mg/dL | MAE = 8.82 mg/dL
ID 42: RMSE = 13.61 mg/dL | MAE = 9.26 mg/dL
ID 43: RMSE = 9.91 mg/dL | MAE = 6.67 mg/dL
ID 44: RMSE = 7.94 mg/dL | MAE = 5.56 mg/dL
ID 45: RMSE = 10.85 mg/dL | MAE = 8.48 mg/dL
ID 46: RMSE = 12.02 mg/dL | MAE = 8.61 mg/dL
ID 47: RMSE = 10.78 mg/dL | MAE = 7.65 mg/dL
ID 48: RMSE = 9.06 mg/dL | MAE = 6.02 mg/dL
ID 49: RMSE = 15.98 mg/dL | MAE = 10.62 mg/dL
ID 50: RMSE = 13.37 mg/dL | MAE = 9.39 mg/dL
ID 51: RMSE = 10.58 mg/dL | MAE = 7.26 mg/dL
ID 52: RMSE = 12.56 mg/dL | MAE = 8.97 mg/dL
ID 53: RMSE = 13.58 mg/dL | MAE = 9.98 mg/dL
ID 54: RMSE = 12.73 mg/dL | MAE = 8.84 mg/dL
ID 55: RMSE = 11.19 mg/dL | MAE = 7.97 mg/dL
ID 56: RMSE = 13.06 mg/dL | MAE = 9.11 mg/dL


The min RMSE is for ID 26 with a value of: 6.51
The max RMSE is for ID 2 with a value of: 23.15
The min MAE  is for ID 26 with a value of: 4.81
The max MAE  is for ID 2 with a value of: 17.47


# Flatten individual model predictions and truths
preds1 = np.concatenate(all_preds1, axis=0)
preds2 = np.concatenate(all_preds2, axis=0)
preds3 = np.concatenate(all_preds3, axis=0)
preds4 = np.concatenate(all_preds4, axis=0)
preds5 = np.concatenate(all_preds5, axis=0)

truths1 = np.concatenate(all_truths1, axis=0)
truths2 = np.concatenate(all_truths2, axis=0)
truths3 = np.concatenate(all_truths3, axis=0)
truths4 = np.concatenate(all_truths4, axis=0)
truths5 = np.concatenate(all_truths5, axis=0)

# Now combine across all models
all_preds_total = np.concatenate([preds1, preds2, preds3, preds4, preds5], axis=0).flatten()
all_truths_total = np.concatenate([truths1, truths2, truths3, truths4, truths5], axis=0).flatten()

# Compute RMSE
combined_mod_rmse = np.sqrt(mean_squared_error(all_truths_total, all_preds_total))
print(f"\nCombined RMSE across all IDs and models: {combined_mod_rmse:.2f} mg/dL")

Combined RMSE across all IDs and models: 12.70 mg/dL


plt.plot(list(range(len(grouped_data['id_1_data']))), list(grouped_data['id_1_data']['gl']))
plt.title('ID 1 (had 2 max RMSE values)')
plt.xlabel('Time index')
plt.ylabel('BG (mg/dL)');


plt.plot(list(range(len(grouped_data['id_2_data']))), list(grouped_data['id_2_data']['gl']))
plt.title('ID 2 (had a max RMSE)')
plt.xlabel('Time index')
plt.ylabel('BG (mg/dL)');


plt.plot(list(range(len(grouped_data['id_20_data']))), list(grouped_data['id_20_data']['gl']))
plt.title('ID 20 (had a max RMSE')
plt.xlabel('Time index')
plt.ylabel('BG (mg/dL)');


plt.plot(list(range(len(grouped_data['id_26_data']))), list(grouped_data['id_26_data']['gl']))
plt.title('ID 26 (had all the min RMSE values, 1 MAE)')
plt.xlabel('Time index')
plt.ylabel('BG (mg/dL)');


plt.plot(list(range(len(grouped_data['id_35_data']))), list(grouped_data['id_35_data']['gl']))
plt.title('ID 35 (had 2 min MAE values)')
plt.xlabel('Time index')
plt.ylabel('BG (mg/dL)');


plt.plot(list(range(len(grouped_data['id_13_data']))), list(grouped_data['id_13_data']['gl']))
plt.title('ID 13 (had 1 max MAE values)')
plt.xlabel('Time index')
plt.ylabel('BG (mg/dL)');


plt.plot(list(range(len(grouped_data['id_18_data']))), list(grouped_data['id_18_data']['gl']))
plt.title('ID 18 (had 1 max MAE values)')
plt.xlabel('Time index')
plt.ylabel('BG (mg/dL)');

	id	time	gl	Age	BMI	A1C	FBG	ogtt.2hr	insulin	hs.CRP	...	freq_low	freq_moderate	freq_severe	glucotype	Height	Weight	Insulin_rate_dd	perc_cgm_prediabetic_range	perc_cgm_diabetic_range	SSPG
0	1636-69-001	2014-02-03 03:42:12	93.0	59.0	21.7	6.7	109.0	205.0	9.0	0.3	...	0.147059	0.369748	0.483193	2	176.3	68.0	0.1015	0.190404	0.026211	91.0
1	1636-69-001	2014-02-03 03:47:12	93.0	59.0	21.7	6.7	109.0	205.0	9.0	0.3	...	0.147059	0.369748	0.483193	2	176.3	68.0	0.1015	0.190404	0.026211	91.0
2	1636-69-001	2014-02-03 03:52:12	93.0	59.0	21.7	6.7	109.0	205.0	9.0	0.3	...	0.147059	0.369748	0.483193	2	176.3	68.0	0.1015	0.190404	0.026211	91.0
3	1636-69-001	2014-02-03 03:57:12	95.0	59.0	21.7	6.7	109.0	205.0	9.0	0.3	...	0.147059	0.369748	0.483193	2	176.3	68.0	0.1015	0.190404	0.026211	91.0
4	1636-69-001	2014-02-03 04:02:12	96.0	59.0	21.7	6.7	109.0	205.0	9.0	0.3	...	0.147059	0.369748	0.483193	2	176.3	68.0	0.1015	0.190404	0.026211	91.0

	id	gl	Age	BMI	A1C	FBG	ogtt.2hr	insulin	hs.CRP	Tchol	...	freq_low	freq_moderate	freq_severe	glucotype	Height	Weight	Insulin_rate_dd	perc_cgm_prediabetic_range	perc_cgm_diabetic_range	SSPG
time
2014-02-03 03:42:12	1636-69-001	93.0	59.0	21.7	6.7	109.0	205.0	9.0	0.3	204.0	...	0.147059	0.369748	0.483193	2	176.3	68.0	0.1015	0.190404	0.026211	91.0
2014-02-03 03:47:12	1636-69-001	93.0	59.0	21.7	6.7	109.0	205.0	9.0	0.3	204.0	...	0.147059	0.369748	0.483193	2	176.3	68.0	0.1015	0.190404	0.026211	91.0
2014-02-03 03:52:12	1636-69-001	93.0	59.0	21.7	6.7	109.0	205.0	9.0	0.3	204.0	...	0.147059	0.369748	0.483193	2	176.3	68.0	0.1015	0.190404	0.026211	91.0
2014-02-03 03:57:12	1636-69-001	95.0	59.0	21.7	6.7	109.0	205.0	9.0	0.3	204.0	...	0.147059	0.369748	0.483193	2	176.3	68.0	0.1015	0.190404	0.026211	91.0
2014-02-03 04:02:12	1636-69-001	96.0	59.0	21.7	6.7	109.0	205.0	9.0	0.3	204.0	...	0.147059	0.369748	0.483193	2	176.3	68.0	0.1015	0.190404	0.026211	91.0

	id	gl	Age	BMI	A1C	FBG	ogtt.2hr	insulin	hs.CRP	Tchol	...	freq_low	freq_moderate	freq_severe	glucotype	Height	Weight	Insulin_rate_dd	perc_cgm_prediabetic_range	perc_cgm_diabetic_range	SSPG
time
2014-02-03 03:42:12	1636-69-001	93.0	59.0	21.7	6.7	109.0	205.0	9.0	0.3	204.0	...	0.147059	0.369748	0.483193	2	176.3	68.0	0.1015	0.190404	0.026211	91.0
2014-02-03 03:47:12	1636-69-001	93.0	59.0	21.7	6.7	109.0	205.0	9.0	0.3	204.0	...	0.147059	0.369748	0.483193	2	176.3	68.0	0.1015	0.190404	0.026211	91.0
2014-02-03 03:52:12	1636-69-001	93.0	59.0	21.7	6.7	109.0	205.0	9.0	0.3	204.0	...	0.147059	0.369748	0.483193	2	176.3	68.0	0.1015	0.190404	0.026211	91.0
2014-02-03 03:57:12	1636-69-001	95.0	59.0	21.7	6.7	109.0	205.0	9.0	0.3	204.0	...	0.147059	0.369748	0.483193	2	176.3	68.0	0.1015	0.190404	0.026211	91.0
2014-02-03 04:02:12	1636-69-001	96.0	59.0	21.7	6.7	109.0	205.0	9.0	0.3	204.0	...	0.147059	0.369748	0.483193	2	176.3	68.0	0.1015	0.190404	0.026211	91.0

	id	gl	Age	BMI	A1C	FBG	ogtt.2hr	insulin	hs.CRP	Tchol	...	freq_low	freq_moderate	freq_severe	glucotype	Height	Weight	Insulin_rate_dd	perc_cgm_prediabetic_range	perc_cgm_diabetic_range	SSPG
time
2016-09-21 00:04:11	2133-004	142.0	54.0	28.09	6.0	145.0	-1.0	11.0	1.6	148.0	...	0.058824	0.117647	0.823529	2	-1.0	-1.0	-1.0	0.155905	0.017363	173.0
2016-09-21 00:09:11	2133-004	142.0	54.0	28.09	6.0	145.0	-1.0	11.0	1.6	148.0	...	0.058824	0.117647	0.823529	2	-1.0	-1.0	-1.0	0.155905	0.017363	173.0
2016-09-21 00:14:11	2133-004	137.0	54.0	28.09	6.0	145.0	-1.0	11.0	1.6	148.0	...	0.058824	0.117647	0.823529	2	-1.0	-1.0	-1.0	0.155905	0.017363	173.0
2016-09-21 00:19:11	2133-004	135.0	54.0	28.09	6.0	145.0	-1.0	11.0	1.6	148.0	...	0.058824	0.117647	0.823529	2	-1.0	-1.0	-1.0	0.155905	0.017363	173.0
2016-09-21 00:24:12	2133-004	132.0	54.0	28.09	6.0	145.0	-1.0	11.0	1.6	148.0	...	0.058824	0.117647	0.823529	2	-1.0	-1.0	-1.0	0.155905	0.017363	173.0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
2016-09-27 04:13:39	2133-004	117.0	54.0	28.09	6.0	145.0	-1.0	11.0	1.6	148.0	...	0.058824	0.117647	0.823529	2	-1.0	-1.0	-1.0	0.155905	0.017363	173.0
2016-09-27 04:18:39	2133-004	117.0	54.0	28.09	6.0	145.0	-1.0	11.0	1.6	148.0	...	0.058824	0.117647	0.823529	2	-1.0	-1.0	-1.0	0.155905	0.017363	173.0
2016-09-27 04:23:39	2133-004	116.0	54.0	28.09	6.0	145.0	-1.0	11.0	1.6	148.0	...	0.058824	0.117647	0.823529	2	-1.0	-1.0	-1.0	0.155905	0.017363	173.0
2016-09-27 04:28:39	2133-004	117.0	54.0	28.09	6.0	145.0	-1.0	11.0	1.6	148.0	...	0.058824	0.117647	0.823529	2	-1.0	-1.0	-1.0	0.155905	0.017363	173.0
2016-09-27 04:33:39	2133-004	118.0	54.0	28.09	6.0	145.0	-1.0	11.0	1.6	148.0	...	0.058824	0.117647	0.823529	2	-1.0	-1.0	-1.0	0.155905	0.017363	173.0

Testing the order of operations on how to build the model before scalling¶

Time to scale the model for more intense use¶

Create a function so it's easier to run multiple iterations¶